context-verify

A command that checks the instruction files used by AI coding assistants, including their length, accuracy, consistency, and freshness.

In plain words
What is it for?
Use it to review context files manually or in CI, with an optional minimum health score.
Why use it?
It finds stale file paths, duplicated instructions, oversized context, and mismatches between shared and tool-specific files.

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/littlebearapps/contextdocs/context-verify
Clone the repo
git clone --depth 1 https://github.com/littlebearapps/contextdocs
Per session 34 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 527 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 $0.00034 $0.00527
Opus 5 $0.00017 $0.00264
Sonnet 5 $0.00007 $0.00105
Haiku 4.5 $0.00003 $0.00053

Measured yesterday against content hash e4ca44f13a4f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

context-verify 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.

commands/context-verify.md · 48 lines

What it actually says

/context-verify

Validate the quality and freshness of AI context files in the current project. Scores signal-to-noise ratio, checks line budgets, detects stale paths, verifies that bridge files stay consistent with AGENTS.md, and flags MEMORY.md drift.

Behaviour

  1. Load the context-verify skill for the full verification framework and scoring rubric
  2. Run all verification checks against existing context files
  3. Calculate and report the context health score

Arguments

  • No arguments: Run full verification with interactive report
  • ci: Output machine-readable format for CI/CD pipelines (exit code 1 on failures)
  • ci --min-score N: Fail if score falls below threshold N

Output

AI Context Health: 80/100 (B — Minor tuning needed)

Breakdown:
  Line Budget:      18/20  (-2 CLAUDE.md bridge restates AGENTS commands)
  Signal Quality:   17/20  (-3 AGENTS.md has file tree)
  Path Accuracy:    16/20  (-2 .cursorrules references src/old.ts, -2 legacy .cursorrules without .cursor/rules/)
  Consistency:      13/15  (-2 CLAUDE.md missing @AGENTS.md)
  Freshness:        12/15  (-3 copilot-instructions.md stale 90+ days)
  Context Load:      4/10  (-3 Cursor 5,400 tokens, -3 Cline 5,200 tokens — over warning)

Checks: 16 run (line budgets, signal quality, stale paths, AGENTS-to-bridge consistency,
MEMORY.md drift, Context Guard status, context load, @import paths, rule path-scopes,
rule symlinks, .mcp.json, agent memory hygiene, plugin manifest, modern Cursor layout,
modern Cline layout, Copilot bridge optionality)

To reach grade A (90+): Migrate to .cursor/rules/agents.mdc (+2), remove file tree (+3),
restore @AGENTS.md import (+2), fix stale path (+2).
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. yesterday First seen · 48 lines · 34 tokens per session scan A e4ca44f13a4f

Subscribe to this mod's changes

context-verify is a command published in the GitHub repository littlebearapps/contextdocs (5 stars, last pushed 1mo ago), licensed MIT. It adds 34 tokens to every session and 527 once invoked, about $0.0002 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 2026-08-31.