ShisaD (shisad)

Security-first AI agent daemon framework.

ShisaD is a long-running daemon that sits between an LLM and external systems (tools, files, network, messaging channels). The model proposes actions; the shared planner execution path decides what actually executes through policy enforcement, provenance/taint checks, confirmation, and audit. Authenticated operator convenience RPCs are separate routes with route-local validation and logging; see the ref-scoped authority map for the exact boundary.

The core question at every action is: who asked for it? ShisaD is the user's agent — it exists to do what the user asks with the highest possible fidelity, and to prevent anything else (prompt injection, hallucination, attacker-controlled input) from taking control.

Rather than ignoring the elephant in the room, our design targets the lethal trifecta head-on: agents that access private data, process untrusted content, and take consequential actions are inherently high-risk. Most agent security research solves this by removing capabilities until the agent is safe but useless. ShisaD takes the opposite approach: keep the agent fully capable and build enforcement infrastructure that makes each capability safe to use at runtime. If a tool is insecure, the goal is to fix the enforcement, not disable the tool.

                       ┌─────────────────┐
                       │   LLM (Planner) │
   Untrusted ─────────►│   [tokens mix]  │─────────► Proposed
   Content             │                 │            Actions
                       └─────────────────┘
                                                       │
   ═══════════════════════════════════════════════════════════
   ║               ARCHITECTURAL BOUNDARY                    ║
   ═══════════════════════════════════════════════════════════
                                                       │
                                                       ▾
                        ┌─────────────────┐     ┌─────────────┐
                        │ Trusted Config  │     │  Security   │
                        │ (policies,      │     │  Analyzers  │
                        │  goals)         │     │ (metadata   │
                        └─────────────────┘     │  only)      │
                                                └──────┬──────┘
                                                       ▾
                                                APPROVE / REJECT

Features

• Structured long-term memory — separate identity, active-attention, recall, procedural, and evidence surfaces share a versioned local store with provenance-gated writes, review gates for high-risk paths, and auditable provenance
• COMMAND/TASK orchestration runtime — persistent COMMAND sessions hand off delegated work to isolated TASK sessions with taint-safe summaries, approval provenance, and explicit task envelopes
• Per-call policy enforcement — the 8-layer PEP pipeline (registry, schema, capability, DLP, resource authorization, egress allowlisting, credential scoping, taint sink enforcement) runs on every shared planner / tool.execute action, not just at session start
• Taint-aware content handling — ingress/egress content firewalls track provenance of untrusted input through the execution path
• Confirmation gates, not blanket denial — user-requested actions proceed; ambiguous or tainted actions route to confirmation; only genuine anomalies trigger lockdown
• Command-channel approvals — supported command surfaces can approve or reject routine pending actions where they originated, with CLI remaining as fallback rather than the only path
• Behavioral anomaly detection — control plane consensus (5 independent voters) for runtime anomaly detection, rate limiting, lockdown escalation, and user-visible warnings on repeated suspicious deny patterns
• Destructive command protection — under the supported containment profile, command policy is enforced before execution rather than by LLM judgment; unavailable required isolation fails closed, while the explicit expert_host_fallback posture carries no supported-containment claim
• Clean-room workflows — admin operations run in a taint-isolated session mode with no auto-apply
• Multi-channel messaging — Matrix, Discord, Telegram, Slack (Socket Mode), with default-deny identity allowlisting per channel
• Assistant primitives — notes/todos, scheduler with shared delivery, web search/fetch, baseline browser automation, filesystem/git helpers, and evidence references for large untrusted output
• Artifact and evidence boundaries — restart-stable evidence refs, structured ArtifactLedger storage, and terminal-safe evidence rendering keep large untrusted content off the raw prompt path by default
• Intent-grounded execution — risky actions must trace back to committed user or clean COMMAND intent, with missing-path reads routed to confirmation and missing-path side effects blocked
• Provider routing — pluggable LLM provider presets (Shisa, OpenAI, OpenRouter, Google, local vLLM) with per-route auth, model selection, and mixed-mode deployment
• Tool-surface integrity — reviewed local skills can declare tools, but their schema hashes are pinned across install/restart/runtime, drifted reviewed tools fail closed with explicit audit visibility, and dynamic remote tool discovery remains fail-closed
• Observability — comprehensive audit trail, TUI dashboard (pending actions, tasks, channel health, alerts), and doctor diagnostics

Status

This repo is public and still pre-alpha. The latest published stable line is v0.8.2. The v0.8.0b1 beta checkpoint remains available for users who need the prerelease structured-authorization checkpoint.

This table tracks major release lines for reader orientation; patch releases like v0.5.1 and v0.5.2 stay in the changelog rather than being listed here.

See docs/ROADMAP.md for more details.

Getting Started

shisad is currently PRE-ALPHA software and probably won't do what you think it will if you're not a developer. The easiest way to get setup is to point Claude Code, OpenAI Codex, or some other strong coding agent to install. When it's more baked, the installation procedure will be better.

Users and agents looking to set up ShisaD on their own system should see docs/DEPLOY.md for the full bring-up guide — host bootstrap, provider configuration, channel setup (Discord, Telegram, Slack, Matrix), and troubleshooting. ShisaD is designed to run on a dedicated instance or container, not inside your development environment.

Quick Start

For v0.8.2 artifacts, install the complete assistant runtime rather than a source checkout:

uv tool install 'shisad[assistant]'
shisad --help
shisad doctor check --component all

The v0.8.2 release can prepare one channel at a time or explicitly publish a combined provider/policy/channel selection. Enroll logical credential references first, then use shisad setup wizard on an interactive unmanaged terminal or shisad setup apply --selection FILE for deterministic automation. The apply command is a dry run unless --write is present. One-channel setup remains available through shisad setup channel --channel <name> ....

The assistant extra contains the Textual UI plus MCP, Matrix E2EE, Discord, Telegram, and Slack client runtimes. It does not enable channels or add credentials; those remain explicit configuration. PromptGuard also remains a separate opt-in, so install shisad[assistant,promptguard] only when that local model runtime is wanted. The latest currently published package is v0.8.2, where the assistant extra is the complete consumer install surface.

This repository also contains a tested Linux/amd64 Dockerfile candidate. No registry image is published or signed from this tree yet; build it locally and follow the volume, policy, and isolation checks in docs/DEPLOY.md.

Development checkout

Use a source checkout for development and test groups:

git clone https://github.com/shisa-ai/shisad.git
cd shisad
uv --no-config sync --frozen --group dev --group channels-runtime

YARA-backed content scanning is included in the base install through textguard[yara]. For local PromptGuard runtime checks from a source checkout, add the security runtime dependency group:

uv --no-config sync --frozen --group security-runtime --group dev --extra chat

--no-config --frozen makes source setup consume the repository's reviewed lock instead of silently replacing it with discovered user/system uv config. Explicit UV_* environment variables and command-line overrides remain caller-owned and should be cleared when they conflict with this setup path. If your environment intentionally enforces a different package-age cutoff, it must admit every security floor in pyproject.toml; do not lower those floors or narrow supported Python versions to make a stale package universe resolve.

For package installs, use the first-class PromptGuard extra, for example uv pip install 'shisad[promptguard]'. security-runtime is a uv dependency group, not a pip extra; use --group security-runtime, not --extra security-runtime. The assistant, chat, and promptguard package sets are project optional extras; assistant is the complete consumer profile, while chat remains the smaller Textual-only profile.

Memory retrieval prefers Python's sqlite3 runtime to have SQLite FTS5 enabled. shisad falls back when FTS5 is unavailable, but you should verify the preferred path with uv run shisad doctor check --component storage; see docs/runbooks/SQLITE.md for install guidance.

Configuration

Environment variables use SHISAD_ prefixes. Full reference: docs/ENV-VARS.md.

The v0.8.2 CLI can inspect the environment and create or inspect the typed TOML surface without starting the daemon:

shisad
shisad init --non-interactive
shisad init --from-env
shisad config validate
shisad config upgrade
shisad config show --format human
shisad config diff --format human
shisad config schema --format json
shisad env --format human
shisad credential status model.primary
shisad start
shisad status
shisad stop

Bare shisad is a read-only welcome and preflight. It distinguishes fresh, returning, explicitly managed, and non-interactive posture from finite machine facts; reports required, optional/degraded, and informational checks; and names an explicit next action. It does not prompt, write configuration, migrate a schema, start the daemon, or open chat. Unsupported config schema and invalid managed posture fail actionably instead of inferring consent. Guided setup is an explicit shisad setup wizard command. After successful interactive publication, its default-exit menu can start the daemon and open chat or the dashboard, or show the guided-tour entry point; managed/non-interactive setup does not prompt or start a process.

shisad start now launches one detached POSIX daemon, waits boundedly for its typed status, and prints only a structural readiness summary plus the owner-only log path. Repeating it against a reachable daemon is idempotent. Use shisad start --foreground for service supervisors, containers, or direct log inspection; unsupported native background platforms receive that foreground path as actionable guidance. shisad stop still shuts down through the control API rather than trusting a PID file.

shisad tour prints a deterministic, maintained walkthrough of readiness, ordinary chat, live policy outcomes, dashboard state, and recovery commands. It consumes the bounded typed O3A health projection when a daemon is reachable and otherwise keeps the guide usable with explicit start/status/doctor recovery. Non-interactive use only prints the guide and that read-only health state. On a TTY, an explicit default-no choice can open the ordinary chat app with a display-only suggested request; the tour never submits that request, invokes a model/tool, changes policy, or creates separate tutorial state.

shisad init creates one commented template at $XDG_CONFIG_HOME/shisad/config.toml (normally ~/.config/shisad/config.toml) and refuses an existing or symlink destination. It enforces owner-only mode on POSIX and reports when equivalent permission tightening is unavailable on another supported platform. It is deliberately not a setup wizard: it does not copy ambient secrets, configure a provider or policy, create daemon state, or start the daemon. Use --non-interactive when scripts want to state the existing no-prompt behavior explicitly. --from-env instead writes only typed, non-default, non-secret fields selected by supported environment variables; secret-bearing fields remain environment-owned and are named only as omitted fields. Use root --config FILE to select another path. Effective precedence remains command-line override, environment, TOML, then typed default; show, diff, env, and validation output redact secret-bearing fields.

Configuration schema 1 is current. shisad config upgrade is a no-write inspection by default. For a legacy config with no schema_version, rerun it with --write to add only schema_version = 1; shisad first preserves the exact original in a .pre-v1.bak file, validates a temporary candidate, and atomically replaces the config. The write summary reports whether owner-only permissions and parent-directory synchronization were supported. Interactive unmanaged startup may apply that one safe migration with the same visible backup and rollback guidance. Managed or non-interactive startup uses the compatible legacy values in memory but does not rewrite the operator file; it prints the explicit upgrade command instead. Newer, malformed, or explicitly older versions are refused without a downgrade attempt.

The O2 credential foundation can keep model-provider secrets out of TOML. A logical reference records only backend metadata; the value stays in an environment variable, an optional OS keyring, or a permission-protected local plaintext file:

# Metadata only: this never copies or prints OPENAI_API_KEY.
shisad credential set model.primary --backend env --locator OPENAI_API_KEY

# Or enroll a local owner-only file value through stdin, never argv.
printf '%s' "$PROVIDER_KEY" |
  shisad credential set model.primary --backend file --stdin

# Point model config at the logical name.
# [model]
# remote_enabled = true
# api_key_ref = "model.primary"

shisad credential status model.primary --format human
shisad credential remove model.primary

The local file backend is plaintext at rest protected by 0700 directories and 0600 files; it is not described as encrypted. Install shisad[credentials] to enable the maintained Python keyring integration. An unavailable keyring fails actionably and never falls back to a file. Generated setup/config output persists references, not values. Existing explicit raw model-key fields remain compatible, but a route cannot configure both a raw key and a reference. A reference supplies a credential; it does not bypass the existing remote_enabled posture.

Credential --stdin normalizes terminal CR/LF input record delimiters before storage. The printf '%s' example avoids adding a shell newline; exact-byte enrollment of values that intentionally end in a newline is not supported.

Provider and policy setup can now be evaluated without writing the final configuration or active policy files:

Authenticated maintained presets require an explicit --credential-ref; vllm_local_default is the only maintained unauthenticated preset and rejects a credential reference. Ambient provider keys do not fill an omitted setup selection.

# Resolves model.primary only in memory, validates the endpoint, and performs
# one bounded planner request. Output contains the logical reference, not its value.
shisad setup provider \
  --preset openai_default \
  --credential-ref model.primary \
  --format human

# Explicitly generate an unverified fragment without making a network request.
shisad setup provider \
  --preset openai_default \
  --credential-ref model.primary \
  --skip-probe \
  --format json

# Generate a validated profile on stdout; no active policy is changed.
shisad setup policy --profile recommended --format human
shisad setup policy --profile strict --format json

recommended keeps default-grant capabilities and automatic per-call handling. strict keeps the same capabilities but requires confirmation by default and requires the semantic classifier and YARA postures. A custom profile requires all three finite choices explicitly (--confirmation, --semantic-classifier, and --yara). Broader rule authoring remains available through the existing policy-file format. A skipped probe is reported as unverified; a failed probe is not retried automatically, and provider response bodies are discarded rather than printed or persisted.

The combined flow accepts the same typed inputs. Interactive setup prompts only when both terminal streams are interactive and managed mode is off; it shows a redacted summary and requires a final confirmation whose default is no. For automation, use one bounded secret-free YAML/JSON selection:

provider:
  preset: openai_default
  model_id: gpt-5.4-2026-03-05
  credential_ref: model.primary
policy:
  profile: recommended
channels:
  - channel: discord
    bot_token_ref: channel.discord
    default_target: "1234567890"
    trusted_users: ["operator-user-id"]

# Evaluate only. No prompt and no write.
shisad setup apply --selection setup.yaml --skip-probes --format human

# Explicitly publish the displayed unverified selection.
shisad setup apply --selection setup.yaml --skip-probes --write --format json

Omit --skip-probes for the bounded live checks. OpenRouter and local vLLM combined selections require an explicit model ID because shisad does not invent a maintained default for those presets. Successful publication creates policy.yaml first and then a schema-validated commented config.toml beside it, both owner-only (0600). Selected fields are active; the rest of the live schema stays commented. Existing or symlink destinations are refused. The two exclusive writes are not a transaction: if config publication fails after the policy completes, the error identifies the inert policy artifact for explicit inspection or removal. Setup never starts the daemon.

Chat, the one-shot terminal dashboard, and the static web snapshot share the three built-in palettes shisa-dark, shisa-light, and shisa-high-contrast. Select one with SHISAD_UI_THEME, disable optional motion with SHISAD_REDUCE_MOTION=true, and suppress palette color with NO_COLOR or a root flag such as shisad --no-color tui. Custom theme-file selection is not a supported configuration surface. In the v0.8.2 release, chat also polls the daemon's canonical action.pending state for the current session and keeps a bounded confirmation panel updated without another prompt. The panel shows only confirmation IDs plus bounded tool, risk, and approval posture, and points to shisad action confirm or shisad action reject for an explicit decision without deciding anything itself. Pending-state RPC or persistence degradation leaves chat input usable and displays a retrying unavailable state instead of treating an unlisted action as resolved. The v0.8.2 release also projects typed action events into a bounded live-progress panel for the exact chat session and one edited Discord message per target/turn. Progress contains only bounded tool identity plus finite lifecycle state—never arguments, values, results, paths, URLs, or error prose—and remains best-effort observability separate from final delivery. An unfiltered shisad events subscribe stream includes both raw audit events and synthetic ActionProgress rows; --count counts every emitted row. shisad web-ui writes a local static investigation/export artifact; it is not the planned live operator web application. The artifact embeds session, pending-action, alert, and egress-review data, so keep it private and remove it when the investigation is complete.

Expected CLI failures use exit status 1 for command/user-state errors, 2 for daemon-connect/RPC errors, a completed unsuccessful setup-provider probe (and Click usage compatibility), and 3 for invalid or unsafe configuration; success is 0. doctor remains read-only and has no automatic --fix mode. Root help advertises the canonical reality-check spelling while the legacy realitycheck spelling remains a hidden compatibility alias.

Recommended: use the runner harness for local development. It handles env isolation, secret loading, and policy bootstrapping:

bash runner/harness.sh start       # background (requires tmux)
bash runner/harness.sh start --fg  # foreground
bash runner/harness.sh status
bash runner/harness.sh shisad status

The runner and plain shisad use the same per-user default socket, so another terminal can run shisad status or shisad chat without exporting SHISAD_SOCKET_PATH. See runner/README.md for details. Secrets go in runner/.env (gitignored) or SHISAD_ENV_FILE.

Manual baseline

export SHISAD_DATA_DIR="$HOME/.local/share/shisad"
case "${XDG_RUNTIME_DIR:-}" in
  /*) export SHISAD_SOCKET_PATH="$XDG_RUNTIME_DIR/shisad/control.sock" ;;
  *) export SHISAD_SOCKET_PATH="/tmp/shisad-$(id -u)/control.sock" ;;
esac
export SHISAD_POLICY_PATH="$PWD/.local/policy.yaml"
export SHISAD_LOG_LEVEL="INFO"

Provider routing

Default (Shisa.AI):

# Planner route remote-enables implicitly when SHISA key resolves.
export SHISA_API_KEY="<shisa-api-key>"

OpenAI:

export SHISAD_MODEL_REMOTE_ENABLED=true
export OPENAI_API_KEY="<openai-api-key>"
export SHISAD_MODEL_PLANNER_PROVIDER_PRESET="openai_default"
export SHISAD_MODEL_PLANNER_MODEL_ID="gpt-5.4-2026-03-05"
# Optional: export SHISAD_MODEL_PLANNER_REQUEST_PARAMETERS='{"max_completion_tokens":512}'

OpenRouter:

export SHISAD_MODEL_REMOTE_ENABLED=true
export OPENROUTER_API_KEY="<openrouter-api-key>"
export SHISAD_MODEL_PLANNER_PROVIDER_PRESET="openrouter_default"
export SHISAD_MODEL_PLANNER_MODEL_ID="qwen/qwen3.5-397b-a17b"
export SHISAD_MODEL_PLANNER_EXTRA_HEADERS='{"HTTP-Referer":"https://example.com","X-Title":"shisad"}'

Google (OpenAI-compatible):

export SHISAD_MODEL_REMOTE_ENABLED=true
export GEMINI_API_KEY="<gemini-api-key>"
export SHISAD_MODEL_PLANNER_PROVIDER_PRESET="google_openai_default"
export SHISAD_MODEL_PLANNER_MODEL_ID="gemini-3.1-pro-preview"

Local vLLM:

export SHISAD_MODEL_PLANNER_PROVIDER_PRESET="vllm_local_default"
export SHISAD_MODEL_PLANNER_BASE_URL="http://127.0.0.1:8000/v1"
export SHISAD_MODEL_PLANNER_REMOTE_ENABLED=true
export SHISAD_MODEL_PLANNER_AUTH_MODE="none"

Mixed mode (planner remote, embeddings local, monitor remote):

export SHISAD_MODEL_REMOTE_ENABLED=true

export SHISAD_MODEL_PLANNER_PROVIDER_PRESET="openrouter_default"
export SHISAD_MODEL_PLANNER_MODEL_ID="qwen/qwen3.5-397b-a17b"
export SHISAD_MODEL_PLANNER_API_KEY="<planner-openrouter-key>"

export SHISAD_MODEL_EMBEDDINGS_PROVIDER_PRESET="vllm_local_default"
export SHISAD_MODEL_EMBEDDINGS_BASE_URL="http://127.0.0.1:8000/v1"
export SHISAD_MODEL_EMBEDDINGS_REMOTE_ENABLED=true
export SHISAD_MODEL_EMBEDDINGS_AUTH_MODE="none"
export SHISAD_MODEL_EMBEDDINGS_MODEL_ID="text-embedding-3-small"

export SHISAD_MODEL_MONITOR_PROVIDER_PRESET="openai_default"
export SHISAD_MODEL_MONITOR_API_KEY="<monitor-openai-key>"
export SHISAD_MODEL_MONITOR_MODEL_ID="gpt-5.4-2026-03-05"

Verify provider setup:

uv run shisad doctor check --component provider

Auth notes:

• Use *_auth_mode=header when custom auth header names are required.
• *_auth_header_name is not accepted for *_auth_mode=bearer|none.

Channels

export SHISAD_DISCORD_ENABLED=true
export SHISAD_DISCORD_BOT_TOKEN="<token>"
# Optional: create/reuse one Discord thread per addressed parent message.
export SHISAD_DISCORD_USE_THREADS=true

export SHISAD_TELEGRAM_ENABLED=true
export SHISAD_TELEGRAM_BOT_TOKEN="<token>"

export SHISAD_SLACK_ENABLED=true
export SHISAD_SLACK_BOT_TOKEN="<xoxb-token>"
export SHISAD_SLACK_APP_TOKEN="<xapp-token>"

# Default-deny allowlist: channel -> [external_user_id]
export SHISAD_CHANNEL_IDENTITY_ALLOWLIST='{"discord":["1234567890"],"telegram":["11111"],"slack":["U12345"]}'

Discord thread mode defaults off. When enabled, parent-channel mentions create a bounded shisad-<message-id> thread, later turns in that thread reuse its session, and replies never fall back to the parent if the thread cannot be resolved. Grant the bot Create Public Threads and Send Messages in Threads.

Assistant surfaces

# web_fetch and web_search are enabled by default.
# web_search needs a compatible JSON search backend (SearxNG-style /search?q=...&format=json).
# IP-literal, localhost, and .local/.internal/.lan backends must be in the effective web allowlist.
# Set SHISAD_WEB_ALLOWED_DOMAINS for that list; if unset, daemon policy egress hosts are used.
# Public backend hosts can be listed to preapprove backend redirects and related fetches.
# For local SearxNG setup, see docs/DEPLOY.md#web-search-backend-recommended.
export SHISAD_WEB_SEARCH_BACKEND_URL="https://search.example.com"
export SHISAD_WEB_ALLOWED_DOMAINS='["search.example.com","docs.example.com"]'

# Verify the configured tool surface from a live daemon:
# uv run python scripts/live_tool_matrix.py --tool-status

# Optional: browser automation baseline (read-mostly navigation plus
# confirmation-gated write actions). The built wheel does not install this
# source-checkout wrapper; package installs need an explicit compatible wrapper.
export SHISAD_BROWSER_ENABLED=true
export SHISAD_BROWSER_COMMAND="/path/to/shisad/scripts/shisad-playwright-cli.mjs"
export SHISAD_BROWSER_ALLOWED_DOMAINS='["example.com"]'

export SHISAD_ASSISTANT_FS_ROOTS='["/tmp/shisad-workspace"]'

Usage

Start and verify

uv run shisad start --foreground

In another shell:

uv run shisad status
uv run shisad doctor check --component all
uv run shisad doctor check --component storage
uv run shisad tui --plain

Sessions

uv run shisad session create --user alice --workspace demo
uv run shisad session list
uv run shisad session message <session-id> "summarize current priorities"

Session-derived conversation summaries can create durable memory entries by default when the session has memory.write; resulting writes are owner-scoped when the session has a complete --user / --workspace owner tuple. For clean demos or workspaces where ordinary chat should not create automatic memory, set SHISAD_MEMORY_AUTO_EXTRACTION_ENABLED=false. Operators can also raise SHISAD_MEMORY_AUTO_EXTRACTION_CONFIDENCE_THRESHOLD from 0 toward 1 to keep only higher-confidence automatic extraction proposals.

Timeline search

uv run shisad memory timeline search "what did we decide last week?" --user alice --workspace demo
uv run shisad memory timeline read <timeline-handle> --user alice --workspace demo
uv run shisad memory timeline promote <timeline-handle> --type fact --key project/decision --user alice --workspace demo

Timeline search is explicit-pull only. Results are archival evidence from prior session transcripts, not current user instructions, and shared/public channel contexts do not reveal owner-private history unless private history is explicitly allowed for that request. Read packets include role, source surface, provenance, taints, evidence refs, and content digests; search/read/promote decisions are recorded in the audit log without raw query or snippet text. For shared-channel contexts, pass the concrete room binding with --recipient, --workspace-hint, and --thread-id so same-connector/different-room history stays isolated.

Notes and todos

uv run shisad note create --key ops/runbook --content "verify doctor before deploy" --user alice --workspace demo
uv run shisad note list --user alice --workspace demo
uv run shisad todo create --title "close rollout checklist" --status open --user alice --workspace demo
uv run shisad todo list --user alice --workspace demo

Web and filesystem

uv run shisad web search "shisad security architecture" --limit 5
uv run shisad web fetch https://example.com
uv run shisad fs read /tmp/shisad-workspace/notes.txt
uv run shisad fs write /tmp/shisad-workspace/out.txt --content "hello" --confirm
uv run shisad git status --repo /tmp/shisad-workspace

Admin clean-room

uv run shisad session mode <session-id> --mode admin_cleanroom
uv run shisad channel pairing-propose --workspace <provider-workspace> --limit 50

Security Model

shisad assumes prompt injection will succeed and builds enforcement outside the model. On the shared planner path, the LLM is a planner, not an executor: it proposes tool calls, while runtime enforcement decides whether each call proceeds, requires confirmation, or is blocked. Those enforcement decisions do not come from the model. Authenticated local operator convenience RPCs do not claim the same PEP/control-plane path; their exact route-local boundaries are listed in docs/AUTHORITY-MAP.md.

The problem: any agent with access to private data (files, email), exposure to untrusted content (web pages, API responses), and the ability to take consequential actions (send messages, write files) is exploitable. This is the lethal trifecta. shisad has all three by design — it's meant to be a useful assistant, not a sandboxed demo.

The approach: instead of removing capabilities until the agent is safe (at which point you've rebuilt ChatGPT with extra steps), shisad keeps all capabilities available and enforces safety per-call:

• 8-layer PEP pipeline on every shared planner / tool.execute action: registry check, schema validation, capability check, DLP (secret pattern detection), resource authorization, egress allowlisting, credential host-scoping, taint sink enforcement
• Taint tracking: content firewalls tag untrusted input on ingress and enforce provenance-aware restrictions on egress — the runtime knows who asked for each action (user vs. injected content vs. model hallucination)
• Confirmation gates: user-requested actions proceed; actions with ambiguous or tainted provenance route to user confirmation with context; only genuine anomalies trigger lockdown
• Approval surface parity: routine channel-originated approvals stay on supported command surfaces with explicit approve/reject affordances or typed channel fallbacks; stronger method-specific proofs route truthfully to browser, helper, or external signer surfaces
• Behavioral anomaly detection: control plane consensus (5 independent voters) catches patterns that individual call-level checks miss
• Destructive command protection: under the supported containment profile, command policy is enforced before execution rather than by LLM judgment; unavailable required isolation fails closed. The separately selected expert_host_fallback posture does not carry that containment claim.

Default posture: built-in tools are not globally removed by a restrictive policy, but configuration-dependent tools remain unavailable until their local or external dependencies and credentials are configured. Operators who need a more restrictive posture deploy an explicit policy via SHISAD_POLICY_PATH.

Shared-path egress model: allowlists auto-approve known-good destinations. Explicit user requests proceed subject to the active policy. Destinations suggested only by untrusted content route through confirmation with warning; unattributed/hallucinated drift is blocked.

See docs/SECURITY.md for the full security architecture and docs/DESIGN-PHILOSOPHY.md for the governing principles.

Architecture

shisad/
├── src/shisad/          # Core source
│   ├── daemon/          # Control API, handlers, runtime implementation
│   ├── security/        # PEP pipeline, content firewalls, taint tracking
│   ├── executors/       # Tool execution, egress proxy
│   ├── channels/        # Matrix, Discord, Telegram, Slack
│   ├── assistant/       # Notes, todos, web, fs/git tools
│   ├── memory/          # Structured storage with semantic search
│   ├── scheduler/       # Task scheduling and delivery
│   ├── cli/             # Click-based CLI
│   ├── ui/              # TUI dashboard
│   ├── skills/          # Hot-reloadable skill plugins
│   └── governance/      # Anomaly voting, consensus
├── tests/
│   ├── unit/            # Component tests
│   ├── integration/     # Cross-component runtime flows
│   ├── behavioral/      # Product-correctness gate
│   └── adversarial/     # Prompt injection, exfil, evasion
├── runner/              # Dev harness (tmux, env isolation, policy bootstrap)
├── scripts/             # Validation, coverage, asset checks
├── docs/                # Design docs, ADRs, runbooks, analysis
└── examples/            # Example configs and skills

Key runtime paths:

• Policy enforcement: src/shisad/security/pep.py
• Egress proxy: src/shisad/executors/proxy.py
• Handler implementation: src/shisad/daemon/handlers/_impl.py (composed from _impl_session.py, _impl_tool_execution.py, _impl_memory.py, etc.)

Development

# Example: validate one affected module and its owning tests.
uv run ruff check src/shisad/core/host_matching.py tests/unit/test_host_matching.py
uv run mypy src/shisad/core/host_matching.py
uv run --python 3.12 pytest tests/unit/test_host_matching.py -q

Ordinary changes use targeted Python 3.12 validation. Full deterministic and multi-version runs are checkpoint/release evidence, and contained suites are not rerun separately. See AGENTS.md for the full validation cadence, coverage rules, and commit conventions.

Documentation

• agentic-security — literature survey on LLM agent security (78 papers, defense taxonomy, production readiness assessment)
• agentic-memory — literature survey on agent memory architectures and poisoning defenses (29+ references, attack taxonomy, defense recommendations)

License

Apache License 2.0. See LICENSE.