Battle-Tested Agent
v1.5.019 production-hardened patterns for AI agents — memory, verification, ambiguity handling, compaction survival, delegation, proof-based handoffs, stale-worker...
⭐ 0· 505·4 current·4 all-time
byDon Zurbrick@zurbrick
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description match the contained assets and the included audit script. Required binaries (bash, grep, find, wc) are exactly what the audit.sh script uses. No extraneous credentials, hosts, or unrelated tools are requested.
Instruction Scope
SKILL.md tells the agent to run scripts/audit.sh against a workspace and optionally copy template assets into the user's workspace. The audit script scans local workspace files (grep/find) for patterns; it does not call external endpoints or attempt to read system-wide secrets. Note: the provided cp snippets will write into the user's workspace and could overwrite files if run without review — review/copy carefully.
Install Mechanism
No install spec — instruction-only with one small shell script. Nothing is downloaded or written to system directories by the skill itself.
Credentials
No environment variables, credentials, or config paths are required. The skill's operations are limited to user-specified workspace directories.
Persistence & Privilege
always:false (normal). The skill does not request permanent agent inclusion or modify other skills; autonomous invocation is allowed by default but this skill's content is passive and agent-initiated.
Assessment
This skill appears internally consistent and low-risk, but take normal precautions before running anything that writes to your files: 1) Inspect scripts/audit.sh locally to confirm behavior (it only greps and checks files). 2) Run the audit against a safe/copy workspace first (e.g., a small test directory) to observe output. 3) Do not blindly run the cp snippets — review and merge templates manually or use cp -n/interactive mode to avoid overwriting important files. 4) Running the audit on very large workspaces may be slow because it recurses with grep; run it against targeted paths if needed. If you need higher confidence, ask the author for a signed release or run the script in a sandboxed environment before applying templates to production workspaces.Like a lobster shell, security has layers — review code before you run it.
battle-testedvk977zphvhxb9yx2ygamy54b8c5823ahwcompactionvk977zphvhxb9yx2ygamy54b8c5823ahwdelegationvk977zphvhxb9yx2ygamy54b8c5823ahwheartbeatvk977zphvhxb9yx2ygamy54b8c5823ahwlatestvk97enwqnf332sjyh21vmy9ac1183s6hsmemoryvk977zphvhxb9yx2ygamy54b8c5823ahwmulti-agentvk977zphvhxb9yx2ygamy54b8c5823ahwproductionvk977zphvhxb9yx2ygamy54b8c5823ahwreliabilityvk977zphvhxb9yx2ygamy54b8c5823ahwsecurityvk977zphvhxb9yx2ygamy54b8c5823ahwself-improvementvk977zphvhxb9yx2ygamy54b8c5823ahw
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
⚔️ Clawdis
Binsbash, grep, find, wc