context

A read-only command that retrieves the architecture decisions governing the files you plan to change.

In plain words
What is it for?
Use it with file paths to find applicable decisions and findings, or without paths to view the queue of proposed decisions awaiting review.
Why use it?
It gives planning work the repository's existing technical context and highlights open proposals or past decisions that should not be silently ignored.

Command

Install

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.

agentmods
npx agentmods add commands/mbeacom/adrkit/context
Clone the repo
git clone --depth 1 https://github.com/mbeacom/adrkit
Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 464 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00018 $0.00464
Opus 5 $0.00009 $0.00232
Sonnet 5 $0.00004 $0.00093
Haiku 4.5 $0.00002 $0.00046

Measured 2d ago against content hash bd0d30d370ed, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

context 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 2d ago.

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.

packages/adapters/spec-kit/commands/context.md · 46 lines

What it actually says

adrkit — governing decisions

Loads the accumulated decision record into context so the plan you are about to write is informed by it, rather than rediscovering it at review time.

User Input

$ARGUMENTS

Steps

  1. Treat any arguments as repo-relative paths the work will touch. Passing the paths you actually intend to change gives a far sharper answer than passing none.
  2. Run {SCRIPT} with those paths as arguments.
    • With paths, it emits a CheckOutcome JSON document: the decisions governing each path, plus findings for any ADR among them.
    • With no paths, it emits the ARB queue as JSON: every proposed decision awaiting review, with its SLA state. That is the set of questions already open — planning against them as if they were settled is the mistake this command exists to prevent.
  3. Read the result and carry it forward:
    • governedBy / governing — decisions already binding here. Follow them, or say plainly in the plan that you are departing from one and why.
    • activeProposals — not yet binding, but in flight. Note any the plan touches; do not silently assume they will land.
    • history — superseded and rejected records. This is where "we tried that" lives, and it is the reason not to re-propose it.
  4. If the command exits non-zero, read the message on stderr. It names the missing thing (no CLI on PATH, no corpus at ADRKIT_DIR) rather than returning an empty result that reads like "nothing governs this."

Notes

  • Read-only. This command never writes to docs/adr/** or anything else.
  • The corpus directory defaults to docs/adr and is overridable with ADRKIT_DIR.
  • An exit code of 1 means the corpus itself has error-severity findings — the report is still complete and still worth reading.
Changes

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.

  1. 2d ago First seen · 46 lines · 18 tokens per session scan A bd0d30d370ed

Subscribe to this mod's changes

context is a command published in the GitHub repository mbeacom/adrkit (11 stars, last pushed 2d ago), licensed Apache-2.0. It adds 18 tokens to every session and 464 once invoked, about $0.0001 per session on Opus 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 2026-08-30.