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/ds1/probe/clarifygit clone --depth 1 https://github.com/ds1/probeWhat 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 | $0.00010 | $0.00415 |
| Opus 5 | $0.00005 | $0.00208 |
| Sonnet 5 | $0.00002 | $0.00083 |
| Haiku 4.5 | $0.00001 | $0.00042 |
Grade A, and why
clarify 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 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.
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.
This is a copy
83% identical to assume — 10 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
What it actually says
Probe: Clarify Thinking
Run the clarify-thinking lens on one input. The lens prompt lives in the clarify-thinking agent; this command only parses arguments and launches it.
Arguments
$ARGUMENTS holds the input and, optionally, an output directory. Parse it in this order:
- If
--out <dir>appears anywhere, that is the output directory. Remove it before the next step. - If what remains starts with a quoted string, the quoted string is the input path. Otherwise, if the first whitespace-delimited token is an existing file, that token is the input path, and a second token (when present and no
--outwas given) is the output directory. - If neither matched, the entire remaining text is the input, to be analyzed directly as pasted text.
Paths that contain spaces must be quoted.
Run
Use the Agent tool with subagent_type: "probe:clarify-thinking". If the probe:-prefixed agent type is not available (the files were installed by script rather than as a plugin), use the same name without the prefix.
Launch prompt:
- Source: the input path, or the full pasted text.
- Output: only if an output directory was given:
<output-dir>/probe-clarify-thinking.md. - Grounding: if the source cites function names, constants, file paths, or schema columns, add: "Verify any specific code claims against the actual code at the cited paths; do not take the source's claims about its own codebase at face value."
Reply
If a file was written, give its path and relay the agent's summary. Otherwise relay the agent's full evaluation.
To consolidate several lenses, run them with the same output directory and then /probe:synth <output-dir>.
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 · 35 lines · 10 tokens per session scan A 8b995531a6f6
clarify is a command published in the GitHub repository ds1/probe (2 stars, last pushed 5d ago), licensed MIT. It adds 10 tokens to every session and 415 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 83% identical to assume, differing in 10 lines, and is treated as a copy.
Other commands, from other repositories
anvil
Start an adversarial debate to stress-test an idea.
anvil-status
Show current Anvil debate status.
anvil-cancel
Cancel an active Anvil debate.
design-review
Workflow recipe — review a design end-to-end, ending in measured numbers rather than adjectives, by chaining 4 skills.
setup-pm-skills
Onboard a new user — find out what they do, recommend the right bundles & top skills, and set up a project CONTEXT.md so every skill is tailored to them.
statusbar-style
Switch the status-bar style (classic / capsule / hairline).