Skip to content
Agentmart
Submit a tool

context-health

Check the health of Persistent Context and prepare the weekly Persistent Context & Controls review: lint instruction files (missing description, broad…

Skill by ALTEN-group

IndexedMIT★ 5
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/alten-group-context-health

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Check the health of Persistent Context and prepare the weekly Persistent Context & Controls review: lint instruction files (missing description, broad applyTo, the everything instruction, near-duplicates, dangling skill/agent/prompt references, secrets, stale files, count growth) and keep an append-only corrections log that proposes new artifacts for topics corrected or re-explained twice. Use when auditing instructions or preparing the review pack.

Use cases: security, marketing, agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found
Runs as
Node.js package

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/context-health

What's inside

Skill name
context-health
Description
Check the health of Persistent Context and prepare the weekly Persistent Context & Controls review: lint instruction files (missing description, broad applyTo, the everything instruction, near-duplicates, dangling skill/agent/prompt references, secrets, stale files, count growth) and keep an append-only corrections log that proposes new artifacts for topics corrected or re-explained twice. Use when auditing instructions or preparing the review pack.
License
MIT
Instructions
49 lines in SKILL.md · view on GitHub

Before installing

  • Ships executable scripts

Files (6)

  • SKILL.md
  • context-health-contract.md · references/
  • context-corrections.mjs · scripts/
  • context-corrections.test.mjs · scripts/
  • context-lint.mjs · scripts/
  • context-lint.test.mjs · scripts/

SKILL.md (first lines)

# Context Health

Keep Persistent Context small, specific, scoped, current, and secret-free, and let evidence decide what to add: a topic corrected or re-explained twice becomes a proposed artifact.

## When to Use This Skill

- The weekly Persistent Context & Controls review needs its pack.
- An instruction file was added or changed and CI should lint it.
- A reviewer corrected or re-explained the same thing again and it should be logged.
- Someone asks whether the instruction set is growing without pruning.

## Path resolution

Resolve `references/` and `scripts/` relative to **this skill's install directory** (the folder that contains this `SKILL.md`), not the consumer repository root or the current shell cwd. If the skill was installed under `.agents/skills/context-health/` or `.claude/skills/context-health/`, use that base path.

## Workflow

1. **Read `references/context-health-contract.md` now.** It lists the rules, severities, the baseline format, and the log format.
2. Lint the consumer repository from this skill's directory:

```bash
node scripts/context-lint.mjs --root . --baseline context-baseline.json
```

Fix errors (missing description, dangling references, secrets). Rotate a reported secret, do not only delete it. Report warnings (long, duplicate, stale, growth) for the Architect to decide.

3. Log a correction or re-explanation as soon as it happens:

```bash
node scripts/context-corrections.mjs add --log specs/learn/corrections.jsonl \
  --topic "Use named exports" --source "PR #123"
```

4. Before the review, list the topics that earned an artifact:

```bash
node scripts/context-corrections.mjs propose --log specs/learn/corrections.jsonl --min 2
```

5. Add control signal-to-noise findings **only as the user supplies them** (how often a control fires and is bypassed). Never invent numbers.
6. `--write-baseline` is a human decision made in a pull request, not something to run to silence the growth warning.

## Done When

- `context-lint.mjs` exits 0 (no errors) on the repository.
- Every proposal comes from the corrections log, and each is presented as a proposal for the Architect.
- No file was changed to hide a finding: no rewritten baseline, no deleted log line.

From ALTEN-group/coding-pal, MIT.

More from ALTEN-group/coding-pal

Signals

Stars
5
Forks
0
Open issues
0
Last commit (this folder)
Oct 13, 2026 (15 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from ALTEN-group

Cite this page

context-health skill. Agent Mart. https://agentmart.dev/skills/alten-group-context-health (last updated 2026-10-24)