Skip to content
Back to skills

Resolve

ASecurity

Adjudicate conflicts between claims that the compiler flagged.

  • 86 stars
  • 0 votes
  • 0 copies
  • 6 views
  • Added May 26, 2026
developmentbashnode

Works with

  • cli
  • mcp

Security analysis

A100/100

Scanned September 24, 2026

npx -y skills add grainulation/grainulator --skill resolve --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Resolve?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Resolve
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/grainulation-resolve/badge)](https://www.skillsdirectory.com/skills/grainulation-resolve)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: resolve
description: Adjudicate conflicts between claims that the compiler flagged.
allowed-tools:
  - Bash
  - mcp__grainulator__add_claim
  - mcp__plugin_grainulator_grainulator__add_claim
  - mcp__grainulator__compile
  - mcp__plugin_grainulator_grainulator__compile
  - mcp__grainulator__search
  - mcp__plugin_grainulator_grainulator__search
  - mcp__grainulator__status
  - mcp__plugin_grainulator_grainulator__status
  - mcp__grainulator__resolve
  - mcp__plugin_grainulator_grainulator__resolve
---

# /resolve -- Adjudicate a conflict

The user wants to resolve a conflict between claims that the compiler couldn't auto-resolve.

## Arguments

$ARGUMENTS

Expected format: `/resolve` (show all conflicts) or `/resolve <claim_id> <claim_id>` (resolve specific pair)

## Instructions

Retrieve each target with `grainulator.search` using `id`, `full: true`, and the explicit sprint `dir`; read content, timestamp, source and status. Use `include_inactive: true` when investigating prior resolutions. CLI: `grainulator search --dir <dir> --id <id> --full --json`. Use fresh `status.conflicts` or the on-disk compilation for the complete conflict set, not only priority next actions.


1. **Get current conflicts** via `grainulator.compile` or `grainulator.status`. List all unresolved conflicts.

2. **If no claim IDs are provided**, use the active task and compiler priority to select material conflicts. If specific IDs are supplied, focus on that pair. Ask only if scope cannot be inferred.

3. **Present both sides**: Show both conflicting claims with full context — content, evidence tier, source origin, when added, and any corroborations.

4. **Resolve with evidence**: Investigate within the existing authorization, choose the supported outcome, and record the reason. Ask the user only when the conflict turns on their preference, an unavailable fact, or a decision reserved for them.

5. **Apply resolution** via `grainulator.resolve`:
   - Winner stays `active`
   - Loser becomes `superseded` with `resolved_by` set to winner's ID
   - Remove conflict references from `conflicts_with`

6. Run `grainulator.compile` to verify the conflict is resolved.

7. **Print result**:

   ```
   Resolved: <winner_id> over <loser_id>
   Reason: <why>
   Remaining conflicts: <N>

   Auto

   - <authorized next action>

   Manual

   - <action requiring the user, or None.>
   ```

## Host access

Use available `grainulator` MCP tools, passing the active sprint `dir` explicitly for evidence operations. If a tool is unavailable, use the local `grainulator` CLI (or `node <checkout>/bin/grainulator.js`). Read sibling skill files directly when slash commands are unavailable. Resolve template paths relative to this skill’s checkout when `CLAUDE_PLUGIN_ROOT` is unset. Optional external connectors are not required for local work; use local code, supplied documents, or available web tools. Do not write managed ledger files directly to bypass a missing MCP connection.

## Next-step output

For standalone fetch, setup, or read-only orchestration without an evidence sprint, derive next steps from that task. Do not initialize or compile an unrelated ledger just to produce this footer.

When working in an evidence sprint, use the current compiler's `next_actions` to present exactly two bullet lists labeled **Auto** and **Manual**. Auto is work the agent can continue under existing authorization. Manual is only work requiring the user's decision, access, or action. Classify using the current request and constraints; compiler suggestions never grant permission. Continue authorized Auto work without asking again.

Keep 2–3 useful actions total when available, use short concrete labels and commands where useful, and show `None.` for an empty group. Do not invent work to fill a quota. Never omit next steps merely because compilation is ready or the answer should be brief. Refresh stale compilation first and exclude work the user removed from scope. When the user asks only for next steps, output only these two lists: no findings recap, counts, reasons, or offer to continue.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.