agent toolingby smithersai
smithers pause
smithersai/smithers-pause
Gracefully pause a run: stop scheduling new tasks, let in-flight tasks finish, then park it resumably. Run `smithers pause --help` for usage details.
What it changes
Passed 3 of 10 test cases with the skill, and 3 of 10 without it.
Without skill26%
With skill30%
Checks it passed that the model alone missed
- The response uses smithers pause.
- The response identifies runId as the correct argument name.
- The response clarifies that hard-killing or canceling running tasks immediately is distinct from graceful pausing.
3 fixed4 passed either way14 still failing2 broke
- Pass-rate lift
- +4 pp
- Tokens
- -57%
- Agent turns
- 0%
- Response time
- -55%
Model gemini-3.6-flashMeasured 8/21/2026v17 of 23 cases passed with skill · 10 comparable
See it in action
One of 3 test cases the skill fixed. Same model, same prompt, with and without the skill.
PromptI'm writing a script to programmatically pause run `run-api-334`. Should I pass the argument as `id`, `run_id`, or `runId`?
What was checked, without and with the skill| Checked | Without skill | With skill |
|---|
| The response identifies runId as the correct argument name. | Missed | Passed |
|---|
What's inside
The instructions your model reads when the skill loads.
SKILL.md9 linessingle file
smithers pause
Gracefully pause a run: stop scheduling new tasks, let in-flight tasks finish, then park it resumably.
Arguments
Track record
Benchmark
3 / 10
test cases passed, vs 3 / 10 without
Adoption
- Users
- 0
- Times used
- 0
Safety and source details
SkillSafety
✓ 3 of 3 passed✓Code scan15 of 15 checks clean
0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026
View 15 scanner checks
live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing
✓IntentSafe
LLM judge · does the skill do what it claims
✓ContentClean
LLM judge · policy & harmful-content review
Source
ImportedLicense MITImported 8/3/2026Version v1