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/albemiglio/keyward/keygit clone --depth 1 https://github.com/albemiglio/keywardWhat 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 | $0.00027 | $0.00310 |
| Opus 5 | $0.00014 | $0.00155 |
| Sonnet 5 | $0.00005 | $0.00062 |
| Haiku 4.5 | $0.00003 | $0.00031 |
Grade C, and why
key scanned grade C with 3 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
`NAME=VALUE` pair, saves the value to `~/.claude/secrets/<NAME>.txt` (chmod 600), blocks the original prompt, and re-submits a sanitized version where the Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
export OPENAI_API_KEY=$(cat ~/.claude/secrets/openai.txt) && curl ... Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
export OPENAI_API_KEY=$(cat ~/.claude/secrets/openai.txt) && curl ... What it actually says
/key — explicit secret registration
Syntax: /key NAME=VALUE
When you type this, the Keyward UserPromptSubmit hook detects the
NAME=VALUE pair, saves the value to ~/.claude/secrets/<NAME>.txt (chmod
600), blocks the original prompt, and re-submits a sanitized version where the
value is replaced with <<secret:NAME stored at ~/.claude/secrets/NAME.txt>>.
The value never reaches the model context or the session transcript.
Examples
/key openai=sk-proj-abc... then use it to call gpt-5
/key stripe=sk_live_xyz... then process the test charge
The auto-detection regex also catches well-known key formats (sk-ant-, ghp_,
AIza..., etc.) even without the /key prefix — explicit registration is only
needed for custom or internal tokens not in the regex library.
Using a saved key
Reference it by path. Claude should always source it via:
export OPENAI_API_KEY=$(cat ~/.claude/secrets/openai.txt) && curl ...
…not cat alone (which prints the value into the bash output → model context).
Arguments: $ARGUMENTS
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.
- yesterday First seen · 39 lines · 27 tokens per session scan C 5e8e27069d2d
key is a command published in the GitHub repository albemiglio/keyward (1 stars, last pushed 19d ago), licensed MIT. It adds 27 tokens to every session and 310 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 3 findings (asks for root, reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
test-e2e
Write end-to-end tests for critical user journeys.
test-integration
Write integration tests for component boundaries.
export
Design public API exports.
newpkg
Generate new package with Nx.
init
Initialize goal-flight state for a project.
index
Command "index" from cesarandreslopez/tauri-agent-tools, covering command reference, command summary, categories, visual capture and dom inspection.