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/capsulegit 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 | $0.01485 |
| Opus 5 | $0.00000 | $0.00743 |
| Sonnet 5 | $0.00000 | $0.00297 |
| Haiku 4.5 | $0.00000 | $0.00148 |
Grade A, and why
capsule 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 — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.
capsule
crabbox capsule captures, replays, and tracks lightweight failure capsules.
A capsule is a portable, replayable failure bundle. Today the capture path is
Actions-first and local-first: crabbox capsule from-actions writes a
capsule.yaml manifest plus bounded local evidence next to it, and the other
subcommands replay, inspect, and annotate that manifest. Replay reuses
crabbox run, so a capsule runs on any lease or environment Crabbox can drive.
A capsule is not a VM snapshot. Use crabbox checkpoint or
crabbox image to save environment state; use crabbox capsule for the failure
source, replay command, oracle, and replay history. See the
Capsules feature guide for the manifest schema and
broader concept.
Subcommands:
from-actions— capture a capsule from a GitHub Actions run.replay— re-run the captured command and record the outcome.inspect— print the source, replay command, oracle, and history.promote— mark the capsule as a regression replay.
A manifest argument may be either the capsule.yaml file or the directory that
contains it; the directory form resolves to capsule.yaml inside it.
from-actions
Capture a failing GitHub Actions run into a capsule.
crabbox capsule from-actions <run-url> --replay '<command>'
crabbox capsule from-actions 123456 --repo example-org/my-app --replay 'go test ./...'
The argument is a GitHub Actions run URL
(https://github.com/owner/repo/actions/runs/123, optionally with
/attempts/N) or a numeric run id. A bare run id requires --repo. --replay
is required: the command is recorded verbatim and is not inferred from workflow
YAML.
Flags:
--repo <owner/name> repository; required when the argument is only a run id
--replay <command> explicit replay command (required)
--output <dir> output directory (default capsules/<owner>-<repo>-actions-<run-id>)
--scenario <text> human-readable scenario (default derived from the run)
--job <name> prefer a specific failed job when a run has several
--required-quality <quality> required replay quality (default semantically_identical)
--max-log-bytes <n> cap retained failed-log bytes (default 262144)
--no-logs skip fetching failed Actions logs
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 · 156 lines · 0 tokens per session scan A 1bc68f12ce42
capsule 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 1,485 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
git
Git operations with intelligent commit messages and workflow optimization.
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.