FD arc-shield
Output sanitization for agent responses - prevents accidental secret leaks
Output sanitization for agent responses - prevents accidental secret leaks
As a process D 48/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions
What is at stake
The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.
The files contain someone else's key or token. If it is live, your agent will call third-party services under a stranger's identity; if it was revoked, the skill's scripts simply fail. Such a key often arrives with the author's whole workspace, personal data included.
The key is visible to everyone who downloaded the skill and has likely been copied by catalog-scanning bots already. Revoke it now, check bills and access logs, then reissue.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The purpose itself is risky: wallets, browser password stores, offensive security. Even an honest implementation gives the agent access to things that cost money.
Explain in the description why the access is needed and how it is limited; add tests that show refusals on dangerous requests.
How to improve
- Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- For Hermes the description must be one sentence under 60 characters; move the conditions to a "When to Use" section.
- 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 · 11
-
high Secrets in code
secret-github-tokenexamples/demo.sh:44GitHub token (test fixture / example file)url = https://ghp_…:x-…@….com/user/repo.git"
fixture -
high Secrets in code
secret-github-tokenINSTALLATION.md:172GitHub token (detector / deny-list definition)echo "Token: ghp_…pqr" | scripts/arc-shield.sh --strict
detector -
high Secrets in code
secret-github-tokenQUICKREF.md:69GitHub token (detector / deny-list definition)echo "Token: ghp_…pqr" | arc-shield.sh --strict
detector
Medium and low: 8
-
medium Secrets in code
secret-private-keyCOMPLETION_SUMMARY.md:102Private key material (detector / deny-list definition; key header without key body; documentation table row)| PGP Key | `-----BEGIN PGP PRIVATE KEY BLOCK-----` | PGP blocks |
detectorheader onlytable -
medium Risky intent
intent-wallet-secretsCOMPLETION_SUMMARY.md:267Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target3. ✅ Wallet recovery phrase when listing file contents
-
medium Secrets in code
secret-private-keyscripts/arc-shield.sh:114Private key material (detector / deny-list definition; key header without key body)echo "$INPUT" | grep -E -- '-----BEGIN PGP PRIVATE KEY BLOCK-----' || true
detectorheader only -
medium Secrets in code
secret-private-keyscripts/output-guard.py:49Private key material (detector / deny-list definition; key header without key body)"PGP Private Key": re.compile(r'-----BEGIN PGP PRIVATE KEY BLOCK-----'),
detectorheader only -
low Secrets in code
secret-aws-keyCOMPLETION_SUMMARY.md:97AWS access key ID (placeholder value)| AWS | `AKIA*` | `AKIA…PLE` |
placeholder -
low Risky intent
intent-wallet-secretsexamples/demo.sh:55Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target (test fixture / example file; quoted — discussed, not commanded)demo_catch "Wallet recovery phrase" \
fixturequoted -
low Secrets in code
secret-aws-keyexamples/demo.sh:60AWS access key ID (placeholder value)"export AWS_ACCESS_KEY_ID=AKIA…PLE
placeholder -
low Secrets in code
secret-high-entropy-tokenSKILL.md:159High-entropy token-like string (may be an id, hash or a credential)kJ8n…hI0 # Novel API key format
Files scanned: 10. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-long-hermesdescription is 74 chars; the Hermes authoring standard requires ≤ 60 (one sentence, ending with a period) - warning
description-no-whenneither description nor a "## When to Use" section says when to use the skill - note
frontmatter-keyunknown frontmatter key "requires"
Process rating: all ten parameters 48/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 2 mutating operations with no state check
- 50Failures and branches. 0 branches, has a failure section
- 60Tools and files. Uses tools (bash, python) that frontmatter does not declare
- 100Steps. 41 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2013 tokens
- low 15 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
- +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 74: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +4Structure: 24 headings
- +3Step-by-step instructions: 41 items
- +4Has examples (11 code blocks)
- +3All 2 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 65.