AC context-architecture
Audit a codebase and bind every claim it makes about itself to a mechanism that fails when the claim stops being true, so it is legible to people and AI agents. Applies Context Architecture's nine principles: make structure say what the system does, place AGENTS.md at boundaries, codify conventions, and bind every claim the repo makes about itself to a mechanism (compiler, linter, automated tests, review) that fails when the claim stops being true. Works greenfield (a repo born legible) and brownfield (a repo restructured in steps). Use when an agent reimplements code that already exists, invents structure, follows stale or deleted docs, propagates a deprecated pattern, or resolves ambiguity at random, or when asked to make a repository "agent-ready", "AI-legible", or to add or fix AGENTS.md / CLAUDE.md files. Related terms: harness engineering, context files, instruction bloat, cognitive debt, agent readiness.
Audit a codebase and bind every claim it makes about itself to a mechanism that fails when the claim stops being true, so it is legible to people and AI agents.
As a process C 58/100 · Has gaps — weak spots: when it triggers, inputs and preconditions, failures and branches
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 1. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
edit-residuethe text marks something as outdated (lines 186): check that old rules are not kept next to new ones — the full check reads the text for contradictions
Process rating: all ten parameters 58/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 60Result and completion. Output format stated, no completion criterion
- 70Execution cost. Instruction body is 4730 tokens
- 85Steps. 28 steps, 1 vague phrases
- 100Tools and files. No external tools needed
- 100Consistency. Name and required fields are in place
- 100Running it twice. Mutating operations check current state
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 10 top-level sections: this looks like several domains in one skill
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 924: 120–800 characters recommended
- +2Single-language instructions
- +5Description quotes 2 example trigger phrases
- +4Structure: 17 headings
- +3Step-by-step instructions: 28 items
- +3Output format is stated explicitly
- +4Has examples (2 code blocks)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 90.