agentleFS
navoditknavoditk/pm-ai-skills-framework/skills/performance-attribution/SKILL.md

performance-attribution

Skill0 starsChanged 44 days agoMITView on GitHub

Explain absolute and relative portfolio performance using authoritative attribution outputs and reconciled contributors.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 6 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use this skill when the user request clearly requires the capability described above.

    line 12
  2. Very specific:

    When activated, read this SKILL.md before selecting tools or making performance claims.

    line 13
  3. Distinctive:

    Do not activate this skill for unrelated market lookup, general education, or requests that can be answered without this business …

    line 18
  4. Distinctive:

    Call approved logical capabilities from the Agentic Data Pipeline.

    line 23
  5. Distinctive:

    Check that the benchmark matches the requested benchmark and that attribution and portfolio snapshots use the same as-of date.

    line 26
  6. Distinctive:

    Prefer authoritative internal analytical capabilities over independent LLM calculations.

    line 34
  7. Distinctive:

    Never fabricate a portfolio value, benchmark, position, return, exposure, or risk statistic.

    line 35
  8. Very specific:

    Use the authoritative performance.attribution capability for official attribution.

    line 44
  9. Distinctive:

    Do not invent residual contributors.

    line 48

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Performance Attribution

Use when

Use this skill when the user request clearly requires the capability described above. When activated, read this SKILL.md before selecting tools or making performance claims. Treat the procedure below as the required workflow.

Do not use when

Do not activate this skill for unrelated market lookup, general education, or requests that can be answered without this business capability.

Procedure

  1. Resolve portfolio, scope, and relevant as-of date(s).
  2. Call approved logical capabilities from the Agentic Data Pipeline.
  3. Validate returned dates, identifiers, and coverage.
  4. Reconcile quantitative claims whenever authoritative totals exist.
  5. Check that the benchmark matches the requested benchmark and that attribution and portfolio snapshots use the same as-of date.
  6. Identify material results rather than narrating every field.
  7. Explain uncertainty or missing data explicitly.
  8. Produce a concise PM-oriented answer grounded in tool results.

Data and tool rules

  • Prefer authoritative internal analytical capabilities over independent LLM calculations.
  • Never fabricate a portfolio value, benchmark, position, return, exposure, or risk statistic.
  • Never bypass authorization.
  • Preserve source and as-of metadata for quantitative claims.
  • In the sandboxed reference evaluation, invoke the staged deterministic bridge with python /workspace/repo/synthetic_data_pipeline/tool_cli.py TOOL_NAME and the relevant flags. Treat its JSON response as the logical-tool result.

Domain rules

  • Use the authoritative performance.attribution capability for official attribution.
  • Reconcile component contributions to total relative return within configured tolerance.
  • Distinguish absolute return from relative return.
  • Include derivative and hedge coverage when positions contain derivatives.
  • Do not invent residual contributors.

Output

Provide:

  • key conclusion;
  • material drivers;
  • supporting quantitative evidence;
  • important limitations/data gaps.

For traceability, end the response with Workflow: performance-attribution after completing the workflow. This marker is an execution trace, not a substitute for the supporting evidence.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.