Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/ai-sdlc-framework/ai-sdlcWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/ai-sdlc-framework/ai-sdlc/context-capture)<a href="https://agentmods.dev/commands/ai-sdlc-framework/ai-sdlc/context-capture"><img src="https://agentmods.dev/badge/commands/ai-sdlc-framework/ai-sdlc/context-capture/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/ai-sdlc-framework/ai-sdlc/context-capture"><img src="https://agentmods.dev/badge/commands/ai-sdlc-framework/ai-sdlc/context-capture.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00036 | $0.00607 |
| Opus 5.5 | $0.00014 | $0.00243 |
| Sonnet 5.5 | $0.00007 | $0.00121 |
| Haiku 4.5 | $0.00004 | $0.00061 |
Grade A, and why
context-capture scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 39 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Capture knowledge into the context engine (RFC-0053). Adapted from the capture-knowledge mechanics of the data-room kit. Capture writes entries, never documents.
Rules (hard)
- Authority is always
inferred. Never writespecialistorcanonical; promotion belongs to the promotion verifier (code) or a human. - Every entry carries a
reverify:note saying how a later reader can re-check it (a file path, a test id, a decision id, or a command to re-run). - Every entry carries a confidence band and a numeric
confidenceinside it:Band Range Meaning candidate below 0.7 weak signal, surfaced as a question working 0.7 to 0.85 agent inference, usable Confidence is capped at 0.85 for agent-written entries. - Every entry carries a source citation and an observed date, and a decay class (volatile 30 days, seasonal 180 days, durable 2 years, evergreen).
- Supersede, never delete. When the user or the source corrects an earlier entry, write a new
inferredentry withsupersedes: <old-id>and mark the old entrysuperseded_by: <new-id>. Do not remove or overwrite the old entry body. - Scope routes the root.
scope: protectedentries go to the gitignored protected root and never into a tracked file or a PR body. - Re-observation of an existing fact moves confidence 30 percent toward the new value and never changes authority.
Steps
- Read the source named in
$ARGUMENTS. - Search for an existing entry on the same fact before writing (
cli-context querywhen available, otherwise grep the knowledge root). - For each new fact, write an entry with the full schema (id, value, confidence, band,
authority: inferred, scope, source, observed, decay, relations,reverify:). - For each correction, follow rule 5 and report the superseded id.
- Print a table: id, band, confidence, supersedes, reverify.
Important
- Running code outranks every written entry; when the source conflicts with the code, record the code's behavior and cite it.
- Never write under a data-room root; capture writes entries only.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 39 lines · 36 tokens per session scan A 4628be562a24
context-capture is a command published in the GitHub repository ai-sdlc-framework/ai-sdlc (428 stars, last pushed today), licensed Apache-2.0. It adds 36 tokens to every session and 607 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-01.
Other commands, from other repositories
compact
Replaces the built-in /compact command and runs automatically on every message.
discover
Discover runtime AI components and static AI dependencies with AIR Blackbox.
replay
Replay recorded AI traces to detect behavioral drift between model versions.
evidence
Export a signed compliance evidence package as JSON or PDF for regulatory handoff.
comply
Run a full EU AI Act compliance scan on your Python AI project.
ai-act-ask
Answer an EU AI Act question grounded in the bundled knowledge base — verbatim statute text, obligation paraphrases, and the compound-risk taxonomy. Offline and deterministic by default; cites the articles it relies on.