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.
git clone --depth 1 https://github.com/yonatangross/orchestkitWrote 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/rules/yonatangross/orchestkit/target-resolution)<a href="https://agentmods.dev/rules/yonatangross/orchestkit/target-resolution"><img src="https://agentmods.dev/badge/rules/yonatangross/orchestkit/target-resolution/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/rules/yonatangross/orchestkit/target-resolution"><img src="https://agentmods.dev/badge/rules/yonatangross/orchestkit/target-resolution.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 | $0.00019 | $0.00667 |
| Opus 5.5 | $0.00008 | $0.00267 |
| Sonnet 5.5 | $0.00004 | $0.00133 |
| Haiku 4.5 | $0.00002 | $0.00067 |
Grade C, and why
target-resolution scanned grade C with 1 finding 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 16d 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.
Tells the agent never to refusehighAnti-refusal
Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.
## Resolve a pronoun target, never refuse it How it starts
The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Resolve a pronoun target, never refuse it
Applies to every skill that takes a target in $ARGUMENTS: assess, verify, implement,
fix-issue, cover, review-pr.
Detect. Strip the flags (--model=, --effort=, --tier= and the like). The target
needs resolving when what is left is a bare pronoun or deictic: them, this, that,
these, those, it, they, same, the above, the last one, that one,
what we just did. An empty target needs resolving only when the skill has no default
for it (review-pr with no argument reviews the current branch's PR; that default stands).
Refusing is the bug, not the safe option. Measured 2026-08-28: /ork:assess them
was sent one message after "bug in orchestkit hooks", mid-investigation of
pretool/bash/dangerous-command-blocker, and the skill replied that "them" had "no
antecedent anywhere in this conversation". It had two (#3780).
Inline skills (the body runs in the main thread)
- Read back for the NEAREST concrete subject: a file just discussed, a diff or PR just opened, a component just investigated, an issue just named.
- Announce the resolution in one line, then proceed with it as the target: "Reading 'them' as the 3 pretool guards we just probed; say otherwise and I'll switch." A wrong guess then costs one correction instead of a turn.
- Ask only when the conversation is genuinely empty (a fresh session that opens with a bare pronoun). That is the one case with nothing to read back to.
Forked skills (context: fork, the body runs in a subagent)
A fork cannot see the conversation, so it cannot read back. Check the target FIRST, before any probe, Agent or Workflow call. If it needs resolving, return exactly this one line and nothing else, then stop:
TARGET_UNRESOLVED: <word>
<word> is the pronoun as typed (this, them), or (empty).
The caller (main thread) on a TARGET_UNRESOLVED result
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.
- 16d ago First seen · 54 lines · 19 tokens per session scan C b78a0cf31b22
target-resolution is a cursor rule published in the GitHub repository yonatangross/orchestkit (298 stars, last pushed today), licensed MIT. It adds 19 tokens to every session and 667 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it C with 1 finding (tells the agent never to refuse). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-12.
Other cursor rules, from other repositories
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
prefer-assertions-over-defensive-checks
Prefer assertions over defensive checks when data is guaranteed to be valid.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.
config-validation-and-normalization
Config validation and normalization responsibilities (defineConfig normalizes, the orm section schema validates and resolves paths, the loader reports per-section diagnostics).