Skip to content

narrative-io/narrative-skills-marketplace / plugins/narrative-common/skills/find-attribute

find-attribute

Verified

Find the canonical Rosetta Stone attribute that best matches a fuzzy description, semantic phrase, or required schema shape. Searches the catalog with pagination, describes the shortlist in one batched call, ranks candidates by name + shape match, and returns the canonical attribute ID plus close alternatives. Use when: "find the X attribute", "what's the graph-edge attribute ID", "look up the email Rosetta Stone attribute", "search the attribute catalog for Y", "which attribute has SOURCE_ID + TARGET_ID + IS_DIRECTED". (narrative-common)

FileKindSize
SKILL.mdskill md15.4 KB
SKILL.md.tmplother13.3 KB
references/EDGE_CASES.mdmarkdown2.0 KB
references/HARNESS_FALLBACK.mdmarkdown2.0 KB
references/HARNESS_FALLBACK.md.tmplother1.4 KB

references/HARNESS_FALLBACK.md

View on GitHub
<!-- AUTO-GENERATED from HARNESS_FALLBACK.md.tmpl — do not edit directly -->
<!-- Regenerate: bun run gen:skill-docs -->
# Harness fallbacks

What to do when `narrative-mcp` is unavailable, when individual MCP
calls error mid-flow, and when the harness doesn't expose
`AskUserQuestion`.

Never silently degrade. If a tool is unavailable, say so explicitly
in the returned `final_answer` (set `confidence: low` and add a
warning).

## `narrative-mcp` unavailable

This skill cannot run — its entire value is searching the live
catalog. The fallback:

- Ask the user for the attribute ID directly. They almost always
  know it (or can copy it from a previous successful run, or pull
  it from `https://api.narrative.io/attributes/<id>` via `curl`).
- If the user provides an ID, return a `final_answer` with
  `attribute_id` set, `confidence: low`, and a warning that the
  schema could not be verified because `narrative-mcp` was
  unavailable.
- If the user can't provide an ID either, return the empty-result
  shape from phase 5 with the warning explaining the harness gap.

## Partial degradation (a single MCP call errors mid-flow)

- `narrative_attributes_search` errors: retry once. If it still
  fails, ask the user for any candidate IDs they remember, then
  jump to phase 3.
- `narrative_attributes_describe` errors: retry once with a smaller
  ID batch (split into halves). If still failing, return the
  search-only ranking with `confidence: low` and a warning that
  schemas were not verified.

## "AskUserQuestion" not available

If the harness does not expose `AskUserQuestion` as a named tool
(Claude Code does; most others don't), ask the user the same question
in plain prose — **one question per turn**, never batched — and wait
for a reply before continuing. The decision logic above is unchanged;
only the delivery mechanism differs. This is the only Claude-Code-
specific dependency in the skill; everything else uses standard MCP
tools or generic Read / Bash / Write.
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall