Official recipe

Reconcile conflicting agent changes with an explicit decision log

Two useful code changes can collide, and accepting one side wholesale may silently discard the other.

Read original recipe

Proposed inputs

Name
Two fictional diffs
Owner
Maintainer
Minimum scope
One shared function plus short intent notes
Quality check
Both diffs use the same base revision.
Name
Expected behavior
Owner
Reviewer
Minimum scope
One assertion for each intended change
Quality check
Assertions cover behavior, not implementation spelling.

Access and setup — proposed or unknown

Proposed minimum: Use a separate working copy and identify the person who can resolve uncertain inputs.
Agree the allowed artifact and review boundary before implementation.
Compatibility unknowns: Unknown: installed runtime, required tools, external service access and present-day compatibility.
External actions requiring approval: Approve repository access, unresolved design choices and any push or shared-branch update.

Detailed steps and output format

At a glance

Suggested result: A reviewable merged patch, intent-to-check map and unresolved-decision list.Suggested; untested.
Tool: Hermes AgentSource-named; check current access.
Suggested inputs: Two fictional diffs; Expected behaviorProposed; access notes below.
Human review: Approve repository access, unresolved design choices and any push or shared-branch update.Suggested boundaries; source approvals below.
Setup notes: Use a separate working copy and identify the person who can resolve uncertain inputs. Agree the allowed artifact and review boundary before implementation.Guide notes; proposed setup untested.
Cost: Workflow cost not established. Check current tool pricing, setup and human review costs before trying it.No measured cost or time added.
Content depth: Source-linked example · suggestions untestedDepth does not establish successful execution.

Check source evidence and limits ↓

Source-backed example. Practical decision guidance is proposed and untested; this is not a full implementation guide.

The problem

Two useful code changes can collide, and accepting one side wholesale may silently discard the other.

Suggested task · untested

Prepare a reconciled change with an explicit explanation of every conflict decision.

What you could produce

A reviewable merged patch, intent-to-check map and unresolved-decision list.

Where these suggestions come from

Core problem: Our suggestion; no result reported by the source
Agent job: Our suggestion; no result reported by the source
Expected output: Our suggestion; no result reported by the source
Relevant job roles: Our suggestion; no result reported by the source
Benefit type: Our suggestion; no result reported by the source

Suggestions and their limits

Describes the proposed purpose and review output; measured benefit and successful delivery are not established.

Editorial synthesis based on the linked source. This framing does not add evidence of successful delivery or results.

Before you try it · suggestions untested

Original decision guidance and first check are proposals. The source supports the attributed task; this guidance has not been executed.

Read the attributed source evidence and original caveats below ↓

Why use this

A merged file can compile while losing required behaviour.
A decision log makes disputed changes easier to review.

Why an agent / simpler alternative

Useful judgment: Reconcile competing intentions without inventing requirements.
Simpler alternative: Use ordinary Git merging for nonconflicting edits.
Unsuitable when: Either change intention is unavailable.

Expected deliverable and review checks

Proposed artifact: A reviewable merged patch, intent-to-check map and unresolved-decision list.
Critical checks: Account for both branch intentions.
Keep unrelated lines unchanged.
Show unresolved choices explicitly.
Missing output evidence: No finished output from an executed recipe was inspected.

First check — proposed, unexecuted

Status: proposed_unexecuted
Synthetic fixture: Two invented branches add input validation and audit logging to the same function; a second conflict contains incompatible timeout choices.
Acceptance checks: Preserve validation and logging together.
Leave the timeout decision unresolved.
List each changed hunk and its reason.
Stop conditions: Stop before repository writes if the base revision or intent is missing.

Tasks: Data & analysis

Useful for: General use

Editorial relevance from the title and source-supported actions; not evidence of the operator’s occupation.

Hermes AgentIndependent testing: No

Failure handling, costs and limits

Failure signals and proposed responses

Signal
One branch’s test disappears
Why it matters
The merge may erase a requirement.
Proposed response
Retain both tests and flag incompatible expectations.
Signal
Both sides change the same policy
Why it matters
Combining values can invent a third policy.
Proposed response
Ask the owner to select the intended rule.

Effort and cost — unknown

Setup work: Prepare scoped inputs and an independent reviewer-owned answer key.
Ongoing review: Recheck changed assumptions against the original acceptance checks.
Unknowns: Unknown: setup effort, usage charges, review time and reliability; no savings established.

Limits

This proposed offline check establishes neither live compatibility nor successful implementation.

Full source evidence

Official recipe · Tool workflows

Original task title: Reconcile conflicting agent changes with an explicit decision log

Editorial candidate. This is source-backed reporting or a documented workflow. BotCurrent has not independently tested this task. A proposed Canadian adaptation is not evidence of a Canadian deployment or compliance certification.

Original source

https://github.com/NousResearch/hermes-agent/blob/main/optional-skills/autonomous-ai-agents/agent-merge-conflict-arbiter/SKILL.md ↗

Attributed to: Hermes Agent

Official recipe

Original action type: tool_action
What the source reports: Reads both branch diffs and stated intents, classifies conflicts, edits conflicted regions, checks behavior and records each merge decision.
Source outcome limits: Recipe only. Build success does not by itself prove both intentions survived. No independent execution test.
Evidence attribution caveat: This official recipe describes a procedure; it does not prove successful execution or independently reproduced results.
Source approval description: Tied or missing intent is escalated. Design choices are exposed for human veto; a routine pre-commit approval gate is not stated.
Named tools: Hermes Agent; agent-merge-conflict-arbiter
Model or version: Skill 1.0.0; agent runtime/model unknown
Publication date: Unknown
Source last checked: 2026-10-14
Demonstrated in Canada: Unknown
Canadian availability: Unknown
Proposed Canadian adaptation: Proposed: review local requirements and approved data scope before any use.
Recommended approval boundaries: Approve repository access, unresolved design choices and any push or shared-branch update.