BC infinity-context
Use when a long agent session is about to hit its context limit, when a detail from an earlier turn must be recalled exactly, or when past sessions should stay searchable offline. Compresses context, archives conversation chunks into a local SQLite/FTS5 store with owner-only permissions and a bounded 30-day retention window, and retrieves exact details on demand. Ships explicitly opt-in maintenance tools that change local files: a retention cleanup that permanently deletes expired archive files (verified archive marker + filename allowlist + --confirm-destructive), an ingest-time retention purge with --purge-only, an INFINITY_CONTEXT_NO_ARCHIVE=1 off switch, and an in-place redaction helper that rewrites a file only inside a declared --allow-dir. Works out of the box on DeepSeek Harness (dsh) and OpenClaw; also runs on Claude Code, Cursor, Dify, Ollama and any Agent Skills host.
Use when a long agent session is about to hit its context limit, when a detail from an earlier turn must be recalled exactly, or when past sessions should…
As a process C 55/100 · Has gaps — weak spots: result and completion, inputs and preconditions, consistency
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill asks for more than the task needs: broad tool access, credential environment variables, binaries. Every extra permission widens the damage from a mistake or a compromise.
Narrow allowed-tools and the variable list to the minimum; replace binaries with readable sources or scripts.
How to improve
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- 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 · 1
✓ No critical or high findings
Medium and low: 1
-
medium Broad scope
meta-broad-allowed-toolsSKILL.md:1Broad tool permissions pre-approved: Bashallowed-tools: Bash Read Write Env
Files scanned: 14. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 5871 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 55/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 19 mutating operations with no state check
- 40Consistency. Frontmatter name (infinity-context) differs from the folder (infinitycontext)
- 50When it triggers. No condition that starts the skill
- 50Failures and branches. 0 branches, has a failure section
- 70Execution cost. Instruction body is 5871 tokens
- 100Tools and files. Tools declared in frontmatter
- 100Steps. 30 steps
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 20 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (9 tags): a typed call is more reliable
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
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 891: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- -213 emoji in the instructions: noise for the model
- +2Single-language instructions
- +4Structure: 26 headings
- +3Step-by-step instructions: 30 items
- +4Has examples (5 code blocks)
- +4Reference files are cited in the instructions (2 of 2)
- +3All 4 scripts are documented
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 77.