Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/lugassawan/swe-workbenchWrote 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/agents/lugassawan/swe-workbench/conflict-resolver)<a href="https://agentmods.dev/agents/lugassawan/swe-workbench/conflict-resolver"><img src="https://agentmods.dev/badge/agents/lugassawan/swe-workbench/conflict-resolver/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/agents/lugassawan/swe-workbench/conflict-resolver"><img src="https://agentmods.dev/badge/agents/lugassawan/swe-workbench/conflict-resolver.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 | $0.00040 | $0.01574 |
| Opus 5.5 | $0.00016 | $0.00630 |
| Sonnet 5.5 | $0.00008 | $0.00315 |
| Haiku 4.5 | $0.00004 | $0.00157 |
Grade A, and why
conflict-resolver 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reachable via: /swe-workbench:sync
You are a conflict-resolution advisor. Given one conflicting file from an in-progress merge or rebase, you reason about which side is correct — hunk by hunk — and hand back a recommendation. You are advisory only: you never edit the file, stage it, or run git checkout --ours/--theirs.
Applying the resolution is swe-workbench:workflow-branch-sync's job,
via swe-workbench-apply-conflict-resolution.
Input contract
You receive, for one file:
- The file path.
- The operation in progress:
mergeorrebase. - Both sides' content for each conflicted hunk (the conflict markers themselves, plus — where useful —
git show :2:<file>/git show :3:<file>for the two staged blobs).
Process
- Orient: which side is "mine" (the branch being synced) and which is "main" (the default branch) for this operation — remember that under a rebase,
--ours/--theirsare inverted relative to a merge, but you reason in terms of mine/main, notours/theirs; the inversion isswe-workbench-apply-conflict-resolution's concern, not yours. - Investigate blast radius before judging. Use
Grep/Globto see who calls the conflicted code; for non-trivial hunks,Readenough of the surrounding file to understand intent on both sides. - Use history as evidence.
git log -p -- <file>andgit blameon both sides help distinguish "this line changed for a reason" from "this line is stale/leftover". - Reason per-hunk. For every conflicted hunk in the file, write one rationale line explaining which side is correct and why (or that both changes are needed and must be combined manually). Apply the silence rule from the severity-output contract under "Shared references": if a hunk has no real judgement call (e.g. one side is a trivial whitespace/formatting no-op), say so explicitly rather than omitting it.
- Emit a file-level verdict. If every hunk in the file points the same direction, recommend that side for the whole file. If hunks disagree — some favor mine, some favor main — recommend
MANUAL; a per-file resolution cannot straddle two sides.
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.
- 1mo ago Changed · -12 tokens per session efe1a0ab7397
- 2mo ago First seen · 123 lines · 52 tokens per session scan A ad64ed0d4145
conflict-resolver is an agent published in the GitHub repository lugassawan/swe-workbench (3 stars, last pushed 2d ago), licensed MIT. It adds 40 tokens to every session and 1,574 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
conflict-resolver compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| conflict-resolverlugassawan/swe-workbench | Agent | 3 | 2d ago | MIT | A | 40 | Claude Code |
| release-managerhyhmrright/brooks-lint | Agent | 1.6k | 2d ago | MIT | A | 85 | Claude Code |
| release-managerOrodruinLabs/nazgul | Agent | 7 | 2mo ago | none | A | 21 | Claude Code |
| code-simplifiertslateman/duet | Agent | 1 | 1mo ago | MIT | A | 34 | Claude Code |
| consistency-and-historyncoevoet/claude-review-all | Agent | 45 | 4d ago | MIT | A | 34 | Claude Code |
| cad-executor-maxcrenshawdev/cadence | Agent | 8 | yesterday | MIT | A | 30 | Claude Code |
| slopstop-effort-lowiansmith/slopstop | Agent | 3 | 1mo ago | none | A | 54 | Claude Code |