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.
npx skills add fending/context-engineering --skill context-aligngit clone --depth 1 https://github.com/fending/context-engineeringWrote 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/skills/fending/context-engineering/context-align)<a href="https://agentmods.dev/skills/fending/context-engineering/context-align"><img src="https://agentmods.dev/badge/skills/fending/context-engineering/context-align/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/skills/fending/context-engineering/context-align"><img src="https://agentmods.dev/badge/skills/fending/context-engineering/context-align.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 | $0.00038 | $0.00941 |
| Opus 5.5 | $0.00015 | $0.00376 |
| Sonnet 5.5 | $0.00008 | $0.00188 |
| Haiku 4.5 | $0.00004 | $0.00094 |
Grade A, and why
context-align 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context Align
Cross-reference your context files against the actual codebase to find drift before it misleads an AI session.
What This Skill Does
Reads all context sources and checks them against the filesystem and project configuration. Checks are ordered by impact:
-
Tech stack references vs dependencies:
- Package names in AGENTS.md vs
package.json/requirements.txt/go.mod/Cargo.toml - Version claims vs actual installed versions
- Framework references (e.g., "React 18" when
package.jsonshows React 19)
- Package names in AGENTS.md vs
-
Directory and file references vs filesystem:
- Paths mentioned in context files that don't exist
- Directory structures described that don't match reality
- References to files that have been moved or renamed
-
Build/test commands vs actual scripts:
- Commands in context files vs
package.jsonscripts /Makefiletargets / CI configs - "Run
npm run test" when the actual script isnpm testorpnpm test
- Commands in context files vs
-
Skill relevance:
- Skill descriptions referencing technologies not in this project's stack
- Skills that reference tools or workflows that no longer exist
-
Cascading contradictions:
Find every AGENTS.md in the project tree (root, subdirectories, global if present). Compare what each level says about the same concern:
- Test runner (Vitest vs Jest vs Mocha, pytest vs unittest)
- Linter and formatter (ESLint vs Biome, Prettier vs project formatter)
- Package manager (npm vs pnpm vs yarn vs bun)
- Framework version (React 18 vs 19, Next.js 14 vs 15)
- Import style (ESM vs CommonJS, aliases vs relative paths)
- Overlapping "Do NOT" rules that widen or narrow the same boundary
When two levels disagree, report both files and what each says. Standard cascading precedence: most specific level wins. If the disagreement looks intentional (legacy module keeping its own test runner), flag it as undocumented -- it should have a comment explaining the exception.
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.
- 1mo ago First seen · 74 lines · 38 tokens per session scan A c0904f874615
context-align is a skill published in the GitHub repository fending/context-engineering (12 stars, last pushed 6mo ago), licensed MIT. It adds 38 tokens to every session and 941 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
ccc
This skill should be used when code search is needed (whether explicitly requested or as part of completing a task), when indexing the codebase after changes, or when the user asks about ccc, cocoindex-code, or the codebase index. Trigger phrases include 'search the codebase', 'find code related to', 'update the…
michel-monitor-pull-request-github-actions
Diagnose a failed, stuck, or never-triggered CI run on a GitHub PR, apply a local fix if possible, push it, and document the result in a single running PR comment. Invoke whenever Michel's CI monitor loop triggers with anyfailure, stuck, or nottriggered — the bash loop already handles pending and allgreen silently, so…
debug
Systematic debugging with persistent state across context resets.
review
Audit code in scope against your architectural rules; report drift (dead code, SoC, monolith, missing-pattern, principle, comment-drift, naming-quality) with propose-confirm per fix; route misses to /capture. Triggers: /review [path | --since | --vault], "review this", "audit the architecture", "check for drift".…
debug
Debugging discipline. Load when the task is fixing a bug, triaging a symptom, or chasing a failing test. Covers reproducing before reasoning, separating symptom from cause, bisecting systematically, not masking what you don't understand, suspecting assumptions, fixing the class not the instance, and leaving a…
incident
Incident response discipline. Load when production is broken, a paging alert is active, or the user is debugging under time pressure. Covers stabilizing before diagnosing, communicating while working, not losing evidence, one change at a time, hypothesizing explicitly, knowing when to escalate, and writing the…