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 skills add ychampion/cskill-agents --skill orphaned-tool-call-sanitizergit clone --depth 1 https://github.com/ychampion/cskill-agentsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ychampion/cskill-agents/orphaned-tool-call-sanitizer)<a href="https://agentmods.dev/skills/ychampion/cskill-agents/orphaned-tool-call-sanitizer"><img src="https://agentmods.dev/badge/skills/ychampion/cskill-agents/orphaned-tool-call-sanitizer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ychampion/cskill-agents/orphaned-tool-call-sanitizer"><img src="https://agentmods.dev/badge/skills/ychampion/cskill-agents/orphaned-tool-call-sanitizer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00027 | $0.00394 |
| Opus 5 | $0.00014 | $0.00197 |
| Sonnet 5 | $0.00005 | $0.00079 |
| Haiku 4.5 | $0.00003 | $0.00039 |
Grade A, and why
orphaned-tool-call-sanitizer 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 7d 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.
What it actually says
SKILL: Orphaned Tool Call Sanitizer
Domain: messaging Trigger: When replaying history for a subagent, make sure assistant tool_use blocks never refer to tool_use_ids that lack results. Source Pattern: Distilled from reviewed subagent orchestration, isolation, and lifecycle implementations.
Core Method
Collect every tool_use_id emitted inside tool_result blocks on user messages, then drop any assistant message whose content includes a tool_use block that references an ID missing from that set. Non-assistant messages and assistant messages without tool_use blocks pass through unchanged, so only the incomplete invocations are pruned before query runs.
Key Rules
- Build the tool use ids with results set exclusively from user-side
tool_resultblocks, since assistant messages can mention tool_use IDs that never resolved. - Treat every assistant message containing a
tool_useentry whoseidis absent from the result set as unsafe and remove it entirely. - Preserve progress, system, and clean assistant messages so the downstream agent still sees the rest of the context.
- Run this sanitizer before providing initial messages to run agent so the async call path never sees an orphaned invocation.
Example Application
When the main agent forks to an Explore or Plan subagent, sanitize the shared history first so the subagent never inherits an assistant tool_use record that was never followed by a tool_result. The downstream query call then succeeds without rejecting the payload.
Anti-Patterns (What NOT to do)
- Forwarding assistant messages with tool_use blocks that lack corresponding results; such payloads trigger API validation errors.
- Waiting until after the query to drop the offending message, which already fails or leaves the transcript poisoned.
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.
- 7d ago First seen · 28 lines · 27 tokens per session scan A 7c0ed65abfd3
orphaned-tool-call-sanitizer is a skill published in the GitHub repository ychampion/cskill-agents (36 stars, last pushed 5mo ago), licensed MIT. It adds 27 tokens to every session and 394 once invoked, about $0.0001 per session on Opus 5. 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 2026-09-03.
Other skills, from other repositories
fix
Dispatch fix subagent for FIX-FIRST gaps from review, re-review, then diagnose unresolved failures after 2 loops.
report
Investigate bugs comprehensively — cascade through trace, capture browser evidence, extract observability data, and prepare or explicitly create a GitHub issue with grounded findings.
trace
Dispatch trace subagent to investigate unknown issues — reproduces, traces, and reports root cause for fix handoff.
perf
Use when auditing performance in any codebase — cold starts, hot paths, dependency weight, storage query patterns. Assess by default, optimize on request; measure, don't guess, and every number carries the command that produced it.
use-ivygrep
Gather focused local repository context with ivygrep before implementing, debugging, reviewing, or explaining code. Use for coding tasks involving unfamiliar paths, branch changes, stack traces, architectural relationships, callers, dependents, tests, configuration, or documentation. Prefer one bounded context pack…
grape
Use Grape MCP for Codex context continuity in coding repositories. Use when a task needs repeated-turn context, omitted context restore, stale-context checks, invalidation checks, or safe continuity across branch and dirty-worktree changes.