ir 0.1.15

An information-retrieval substrate for agentic systems

ir

An information-retrieval substrate for agentic systems — one uniform "find the relevant things in this corpus" contract that scales from an ad-hoc search over an ephemeral list to a maintained capability-discovery engine.

Give an agent one search tool, not fifty tool schemas. ir retrieves candidates, commits to a small high-precision subset (the distractor problem is the central selection risk — fewer, better candidates beat more), and discloses each committed item's payload only when asked.

import ir

# Define a corpus, build the index (incremental), then discover:
source = ir.CorpusSource.from_skills()       # or from_packages(), from_md_reports(), from_files(...)
corpus = ir.build(source)                     # embed + persist under XDG dirs
result = ir.discover(corpus, "how do I deploy the app to the server")

for item in result.results:
    print(item.score, item.name)              # the committed few (or result.abstained)
print(result.to_dict())                       # JSON-serializable (qh / HTTP ready)

Install

pip install ir

ir is light by default — numpy + dol for storage, plus ef / vd for embedding and lexical/hybrid retrieval. Python ≥ 3.10.

Notes for the default (semantic) path:

• The default embedder is all-MiniLM-L6-v2 (384-dim, via sentence-transformers), downloaded on first build (needs network) and cached under ~/.cache/ir. For a fast, offline, dependency-light run — tests, CI, quick experiments — pass embedder="light" (a numpy-only hashing embedder): ir.build(source, embedder="light").
• ir sets USE_TF=0 on import so transformers does not pull in TensorFlow (which crashes on some numpy ABIs); import ir before anything that imports transformers.
• Case generation (ir.eval_gen) and the optional LLM selector need an LLM via oa — install the extra, pip install "ir[llm]". Scoring and evaluation themselves stay offline.

The pipeline

ir is a five-stage pipeline, each stage a small, swappable seam:

Stage	Entry point	What it does
source	CorpusSource	what is in the corpus + what counts as stale
index	ir.build	decompose artifacts into embeddable surfaces, embed, persist (incremental, idempotent)
retrieve	ir.search	hard metadata filter + dense / lexical / hybrid ranking
select	ir.select	commit to a distractor-robust subset, or abstain
disclose	ir.disclose	load the heavy payload (SKILL.md body, package pointer, file text) for committed items — append-only

ir.discover chains retrieve → select → disclose into the single agent-callable (and qh-exposable) tool.

Retrieve

hits = ir.search(corpus, "deploy app", mode="hybrid")   # dense | lexical | hybrid (RRF)

Dense is exact brute-force cosine; lexical is Okapi BM25; hybrid fuses both by Reciprocal Rank Fusion (the strongest default for short, identifier-heavy capability text). Lexical/hybrid reuse vd; dense needs only numpy.

Hybrid has a second fusion, fusion="blend" — a magnitude-preserving score blend instead of rank-RRF. RRF discards score magnitude, which is exactly what abstention calibration needs, so blend separates in-scope from out-of-scope queries far better (and even beats dense); the tradeoff is lower lexical recall on terse corpora, so RRF stays the default. Use blend when abstention matters — see ir_08.

Select

sel = ir.select(hits)                      # conservative default: stay within rel of top, cap at max_k
sel = ir.select(hits, min_score=0.4)       # opt in to abstention ("nothing applies")
sel = ir.select(hits, strategy="score_gap")  # elbow cut, or "top_k" / "rel_threshold" / a callable

The abstention floor is mode-specific (dense cosine, BM25, and RRF live on different scales), so rather than guess min_score, calibrate it from a case file and let discover load it:

ev.calibrate_min_score(corpus, cases, mode="dense", persist=True)  # learn + store the floor
ir.discover(corpus, query, mode="dense", min_score="auto")         # abstain by the calibrated floor

Calibration separates in-scope from out-of-scope query top-scores and picks the floor that best splits them — see ir_07; it works best on dense / lexical (hybrid's RRF scores barely separate). min_score defaults to None (never abstain), so abstention stays fully opt-in.

The conservative defaults (max_k=3, rel=0.9) are tuned, not guessed — see ir_06; re-tune for your own corpus with ev.sweep_selector / ir sweep-select.

Selection is relative (ratios to the top score), so one selector works across dense / hybrid / lexical whose absolute scales differ by orders of magnitude. The result carries auditable signals and a reason — no opaque "confidence" float. An optional LLM selector (make_llm_selector, lazy on oa, injectable for tests) falls back to the heuristic on any failure.

Disclose

payloads = ir.disclose(sel, level="body")  # "metadata" (no I/O) | "body" | "bundled"

Disclosure is a pure read that follows the pointer already stored on each hit (skill_path / path); it never mutates the ranked hits and tolerates a stale pointer. Keeping the agent's context append-only (to protect the prompt cache) is then the caller's discipline — ir hands back additive payloads.

Evaluation

ir.eval scores discovery quality offline (reusing ef's retrieval metrics):

from ir import eval as ev

cases = ev.load_cases("skills_eval.jsonl")               # query + gold artifact_id(s)
ev.evaluate_discovery(corpus, cases, mode="hybrid")      # recall@k / NDCG@k / MRR / MAP + failure taxonomy
ev.evaluate_selection(corpus, cases, strategy="conservative")  # conditional commit rate + selection P/R/F1
ev.sweep_selector(corpus, cases)                         # tune max_k × rel; .best() / .frontier() / .table()
ev.distractor_robustness_curve(source.scope, probes)     # accuracy vs catalog size

evaluate_selection's headline is the conditional commit rate — the selection decision isolated from retrieval (did the selector keep the gold, given retrieval surfaced it?). sweep_selector scores a whole max_k × rel grid against the cases off one retrieval pass, so the selector defaults can be read off the data (.best()) rather than guessed. Generate cases by back-translation with ir.eval_gen (needs an LLM; scoring stays offline).

CLI

ir build skills                          # build/update a preset corpus
ir search skills "deploy the app"        # rank candidates (retrieval only)
ir discover skills "deploy the app"      # retrieve -> select
ir discover skills "deploy the app" --disclose       # + load bodies
ir discover skills "deploy the app" --min-score auto # + calibrated abstention
ir ls                                    # list corpora + record counts
ir info skills                           # config, stats, calibrated floors
ir register notes files --root ~/notes --pattern '.*\.md$'  # register a custom corpus
ir rm notes                              # unregister (keeps built data)
ir eval-gen skills skills_eval.jsonl     # generate eval cases (needs oa/LLM)
ir eval skills skills_eval.jsonl         # score retrieval on a case file
ir eval-select skills skills_eval.jsonl  # score the selection stage
ir sweep-select skills skills_eval.jsonl # tune the selector (max_k × rel) on your corpus
ir calibrate-min-score skills skills_eval.jsonl --persist  # calibrate the abstention floor

Design

The design is grounded in a set of capability-discovery research reports and eval-run findings under misc/docs/ (ir_01–ir_08): the single-search-tool pattern, indexing & embedding strategy, evaluation, the ef + vd reuse analysis, the dense-vs-lexical-vs-hybrid eval, selector tuning, abstention-floor calibration, and magnitude-preserving fusion. ir is light by default (numpy / dol) and reuses the ecosystem (ef, vd, oa) only where it composes cleanly.