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 agentmods add commands/openclaw/crabbox/verifygit clone --depth 1 https://github.com/openclaw/crabboxWhat 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 | $0.00000 | $0.00771 |
| Opus 5 | $0.00000 | $0.00385 |
| Sonnet 5 | $0.00000 | $0.00154 |
| Haiku 4.5 | $0.00000 | $0.00077 |
Grade A, and why
verify 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 2d 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 — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
verify
crabbox verify checks a signed run receipt produced by crabbox run --attest.
It recomputes the receipt's canonical bytes, verifies the embedded Ed25519
signature, and prints the signer's public key fingerprint, so a reviewer can
confirm a pasted receipt has not been edited after the run once they have
matched the fingerprint through a trusted channel.
crabbox run --attest receipt.json -- go test ./...
crabbox verify receipt.json
On success it prints one line and exits 0:
PASS receipt.json signer=sha256:6a5f... trust=self-signed
If any signed field was modified after signing, it prints a failure line and exits 1:
FAIL receipt.json signer=sha256:6a5f... trust=self-signed: signature mismatch
A receipt that cannot be checked at all (unreadable file, invalid JSON,
duplicate JSON keys, a missing or malformed public_key or signature) exits
2 with an error on stderr. Duplicate keys are rejected outright because JSON
parsers disagree on which duplicate wins, which would let an edited receipt
show one value to a reader while the signature still verifies against another.
Receipt format
Schema v1 is the existing successful-run format. It is a flat JSON object.
schema_version, generated_at,
provider, command, exit_code, command_ms, and public_key are always
present; lease_id, slug, run_id, actions_url, and log_sha256 appear
when the run produced them. log_sha256 is the SHA-256 of the combined
stdout and stderr stream as observed by the client during SSH-backed runs.
The signature covers the canonical encoding of every field except signature
itself: the object is re-marshaled as compact JSON with lexicographically
sorted keys, and the Ed25519 signature is computed over those bytes. Because
public_key is inside the signed payload, swapping in a different key also
fails verification.
Schema v2 terminal receipts are emitted for non-zero local runs and stored by the coordinator for brokered terminal runs. They additionally bind the run, lease, slug, raw command digest, final exit code, sync/command/total timing, timestamps, retained log hash, truncation state, full observed stream hash, and signer fingerprint. Their signature input is a fixed length-prefixed field sequence shared by the Go CLI and Worker verifier. Command display text is bounded; unusually long displays are replaced by a marker containing the exact raw-argument digest.
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.
- 2d ago First seen · 75 lines · 0 tokens per session scan A a30cc9c03a08
verify is a command published in the GitHub repository openclaw/crabbox (1,343 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 771 tokens. 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 commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.