Borrowing it
Nothing to install: this file belongs to pvliesdonk/paperless-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/pvliesdonk/paperless-mcp/main/.agents/skills/designing-tool-outcomes/SKILL.mdgit clone --depth 1 https://github.com/pvliesdonk/paperless-mcpWrote 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/pvliesdonk/paperless-mcp/designing-tool-outcomes)<a href="https://agentmods.dev/skills/pvliesdonk/paperless-mcp/designing-tool-outcomes"><img src="https://agentmods.dev/badge/skills/pvliesdonk/paperless-mcp/designing-tool-outcomes/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/pvliesdonk/paperless-mcp/designing-tool-outcomes"><img src="https://agentmods.dev/badge/skills/pvliesdonk/paperless-mcp/designing-tool-outcomes.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 | $0.00070 | $0.02213 |
| Opus 5.5 | $0.00028 | $0.00885 |
| Sonnet 5.5 | $0.00014 | $0.00443 |
| Haiku 4.5 | $0.00007 | $0.00221 |
Grade A, and why
designing-tool-outcomes 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 25d 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.
This is a copy
100% identical to designing-tool-outcomes — 0 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.
How it starts
The opening of the file, as written. The whole thing — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Designing tool outcomes
A tool call ends in one of four outcomes. Classify each case before you write
its code. The class fixes three things at once: what the model receives, what
the message tells it to do next, and the log level. The evidence behind every
rule is in docs/design/reference/mcp-tool-outcomes-and-errors.md and
docs/design/reference/negative-outcomes-and-faults.md.
The four outcomes
The contract is what the tool's description and return type promise. An
outcome that fits the contract is a result. An outcome that cannot produce the
promised thing is an error, however ordinary it is. The two texts split the
work: the description states the contract, limits that hold for every call
included, and never lists failures (writing-model-facing-text, "The
contract, not the failures"). So the error text is the one place a failure
is explained, and it has to stand on its own.
| # | Outcome | Examples | Code | The message tells the model | Log |
|---|---|---|---|---|---|
| 1 | Contract met | the note's text; an empty list from a search that ran to completion | return the value |
nothing extra | the middleware's completion line |
| 2 | Change the request | no note at that path; invalid input; path outside the store; write refused until if_match is passed; a feature this deployment leaves off (a read-only vault, an opt-in feature not enabled) |
raise ToolError(msg, log_level=logging.INFO) |
what was wrong and what to call or pass instead; repeating the same call will not help | INFO |
| 3 | Refresh, then retry | stale version or etag; conflicting concurrent change | raise ToolError(msg, log_level=logging.INFO) |
read again, reapply the change, retry with the new token | INFO |
| 4 | The server failed | I/O error; permissions on the server's own files; a stored file the server cannot parse; index or dependency unavailable; required configuration left out; a bug | let it reach the boundary, or raise ToolError(msg) after logging |
the request was fine; retry later (transient) or tell the user (needs an operator); do not change strategy | WARNING if it heals itself, else ERROR with the traceback |
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.
- 25d ago First seen · 152 lines · 70 tokens per session scan A c11df839a833
designing-tool-outcomes is a skill published in the GitHub repository pvliesdonk/paperless-mcp (1 stars, last pushed yesterday), licensed MIT. It adds 70 tokens to every session and 2,213 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 100% identical to designing-tool-outcomes, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
triaging-incidents
Creating, triaging, updating, and analyzing ServiceNow incidents, including trend reports. Assess severity, check affected CIs, find related incidents, and recommend assignments. Use when the user mentions incidents, INC numbers, outages, service disruptions, ticket creation, triage, priority, severity, ITSM…
xcodebuildmcp-cli
Official skill for the XcodeBuildMCP CLI. Use when doing iOS/macOS/watchOS/tvOS/visionOS work (build, test, run, debug, log, UI automation).
triage
Diagnose and fix broken MCP servers in mcptoon — doctor, health, per-server probes, and the common failure playbook.
opik-diagnose
Surface the Opik traces worth a developer's attention, ranked by signal — Diagnostics issues first, then errors, failed tool calls, latency, regressions, and low online-eval scores. With the Opik MCP connected it lists the project's agentinsightsissue entities, offers to turn Diagnostics on when the project has it…
opik-explain
Root-cause a specific Opik trace, or a pattern across traces, and return a grounded explanation. Uses the hosted Opik MCP when it is connected, and falls back to SDK scripting otherwise. Returns the root cause, the evidence spans as clickable Opik UI links, and one suggested next step. Use for "why did this trace…
debug-failed-job
Find out why a Keboola job failed and propose a fix. Use when the user says a Keboola job, extractor, transformation, writer or flow failed or errored, or asks why their pipeline is red.