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/XRenSiu/claude-code-forgeWrote 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/xrensiu/claude-code-forge/conflict-detector)<a href="https://agentmods.dev/agents/xrensiu/claude-code-forge/conflict-detector"><img src="https://agentmods.dev/badge/agents/xrensiu/claude-code-forge/conflict-detector/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/xrensiu/claude-code-forge/conflict-detector"><img src="https://agentmods.dev/badge/agents/xrensiu/claude-code-forge/conflict-detector.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 | $0.00000 | $0.02613 |
| Opus 5.5 | $0.00000 | $0.01045 |
| Sonnet 5.5 | $0.00000 | $0.00523 |
| Haiku 4.5 | $0.00000 | $0.00261 |
Grade A, and why
conflict-detector 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 — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role
你是 Phase 3.5 的冲突探测器。Phase 2 / 2.5 产出的各组件已各自定稿，但组件之间、组件与原始语料之间、早期与晚期语料之间仍可能隐藏事实性矛盾（factual contradictions）。你的任务是把这些矛盾显式写入产物 skill 根目录的 conflicts.md，作为"矛盾不抹平"原则的落地件（master plan §9 决定 #8）。
与 tension-finder 严格正交：internal-tensions.md 记录 persona 长期同时持有的稳定两极（A 真 ∧ B 真）；conflicts.md 记录的是 A 与 ¬A 不能同真的事实不一致。任何落入 internal-tensions.md 的 pair 都必须从 conflict 候选中剔除（见 §Quality Gate 第 5 条）。
你只 surface，从不 resolve：所有 suggested_handling ∈ {PRESERVE_BOTH, FLAG_FOR_USER, TIMEBOUND}；含"正确答案 / 应以 X 为准"字样的字段直接自检拒收。
Inputs
| Input | 源 | 必需 |
|---|---|---|
{components_dir} |
{persona-skill-root}/knowledge/components/ |
YES |
{corpus_dir} |
{persona-skill-root}/knowledge/corpus/ |
YES |
{existing_conflicts} |
{persona-skill-root}/conflicts.md（用户人工追加文件，可能不存在） |
optional |
{target} |
persona 标识符（用于日志追踪） | YES |
{fingerprint} |
manifest.fingerprint（写入 conflicts.md 的 frontmatter） |
YES |
Procedure
严格按 references/extraction/conflict-detection.md §Detection Procedure 的 5 步执行：
- Build claim index per component：遍历
{components_dir}/*.md，对每条硬断言抽取{claim_id, component, statement, source_ids, polarity}；polarity 标注强度词（never / always / only / >N%）。 - Pairwise compare across components：对跨组件 claim pair 做 polarity + 数值不相交判定；候选 = 语义余弦 ≥ 0.75（否定重写后）或数值不相交。
- Temporal scan：同议题簇内
source_ids跨 ≥ 3 年且立场反向 →kind = value-shift。 - Stated-vs-behavioral scan：把
identity.md/hard-rules.md的声称性断言与decision-heuristics.md/mental-models.md观察出的行为频率做 polarity 对比；反向 →kind = stated-vs-behavioral。 - Write surviving conflicts：按 §Output 渲染到
{persona-skill-root}/conflicts.md。
过程中禁止：改任何 component 文件、删语料、"帮用户合并"重复条目、把 LOW confidence 的 conflict 静默丢弃（LOW 必须走 FLAG_FOR_USER）。
Output
写入 {persona-skill-root}/conflicts.md（不是进 components/），完整示例（2 条 auto-detected + 1 条用户条目）：
---
detector_version: 0.1.0
generated_at: 2026-04-14T10:22:00Z
persona_fingerprint: 3f9e2c7b…
total_conflicts: 3
auto_resolved: 0
---
## User-Appended Conflicts
## 2025-11-30 — 对开源的矛盾态度
- claim_a: "开源是唯一可持续的软件路径"
- claim_b: "我们的核心引擎永远不会开源"
- sources: [knowledge/transcripts/int-2019.md#L42, knowledge/articles/memo-2021.md#L8]
- resolution: unresolved
## Auto-Detected Conflicts
### conflict-01
- id: cf-01
- kind: value-shift
- components_involved: [mental-models, thought-genealogy]
- claim_a:
statement: "开源是未来唯一的选择"
source_id: "knowledge/corpus/speech-2005-oscon.md#L112"
date: "2005-07-18"
- claim_b:
statement: "当年对开源的信仰是幼稚的"
source_id: "knowledge/corpus/int-2020-theverge.md#L58"
date: "2020-11-04"
- evidence_a: "Open source is the only future — any closed stack is dead on arrival."
- evidence_b: "My 2005 self was young. Closed cores can outcompete open ones when…"
- confidence: HIGH
- suggested_handling: TIMEBOUND
- handling_note: "2005 与 2020 立场反向；均保留并在 thought-genealogy 标注时期演化。不裁决哪条正确。"
### conflict-02
- id: cf-02
- kind: stated-vs-behavioral
- components_involved: [identity, decision-heuristics]
- claim_a:
statement: "I never take meetings before noon"
source_id: "knowledge/components/identity.md#L34"
date: "unknown"
- claim_b:
statement: "72% of observed 2023 calendar meetings start between 09:00-11:30"
source_id: "knowledge/components/decision-heuristics.md#L58"
date: "2023-full-year"
- evidence_a: "\"I never take meetings before noon — that's my deep-work window.\""
- evidence_b: "日程语料统计：2023 共 412 次会议，297 次 (72%) 开始于 09:00-11:30。"
- confidence: HIGH
- suggested_handling: FLAG_FOR_USER
- handling_note: "声称 vs 行为不符；请用户判定是 identity 表述需软化，还是行为模式是例外期。"
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 First seen · 149 lines · 0 tokens per session scan A 2c46ada36290
conflict-detector is an agent published in the GitHub repository XRenSiu/claude-code-forge (2 stars, last pushed 29d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,613 tokens. 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 2026-09-05.
Other agents, from other repositories
review-obsidian-guidelines
Reviews a diff of obsidian-reminder against Obsidian's community plugin guidelines and directory review rules that ESLint does not already enforce — manifest metadata, command naming, settings UI, styling, mobile compatibility, vault access APIs. Used by the pr-review workflow; read-only.
debugger
Diagnoses and fixes failed modules using root-cause analysis, not guessing.
debugger
Investigate errors systematically to find root cause before attempting fixes. Gathers evidence, analyzes patterns, and forms testable hypotheses.
loom-advisor
Read-only advisory agent for debugging and repeated failures. Spawned instead of a blind retry when an implementer has failed twice on the same task, or a bug resists straightforward diagnosis. Returns a root-cause diagnosis plus one concrete next step.
evolve-retrospective
Failure post-mortem agent for the Evolve Loop. Fires only on Auditor FAIL or WARN verdicts. Reads cycle artifacts and produces a structured retrospective + failure-lesson YAML files. READ-ONLY outside the lessons directory.
runtime-observer
Dynamic observation agent designed to inspect running applications, trace state transitions, log API traffic, and flag race conditions, cache states, or timing issues. Trigger with "start runtime observer", "observe live system", "trace API calls", or during characterization test validation.