osiman/reader
1
0
Fork 0
Code Issues 1 Pull Requests Actions Packages Projects Releases Wiki Activity
69 Commits 1 Branch 0 Tags
8e27cca16643a174e9e1bb2856d8ff8a24a49f8b
Commit Graph
69 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
root 8e27cca166 docs: sync reader-digest-flow skill with Hermes version (absolute path, extract_failed handling, rerun include_read) 2026-08-03 09:55:19 +08:00
zhuyongxin 807027976e docs: update digest template and localize references 2026-07-29 10:20:20 +08:00
zhuyongxin 6dd8cef347 refactor: simplify reader digest skill 2026-07-28 19:14:09 +08:00
root 5eb390e3ed docs: 添加 Agent Skill 到项目仓库 
- 复制 reader-digest-flow skill 到 skills/ 目录（含 SKILL.md + references/）
- README 新增 Agent Skill 章节说明供 Agent 使用的工作流
 2026-07-28 18:38:54 +08:00
root 7b791ac947 docs: 重写 README 为项目介绍 
从 MCP 集成技术文档改为完整项目 README，包含：
- 项目定位与核心能力
- 架构概览（流程图 + MCP 工具层 + CLI）
- 快速开始与环境变量
- 项目结构一览
- 关键技术决策说明
- 关键词治理流程
 2026-07-28 16:26:34 +08:00
root cdbcdcd485 feat: pipeline 并行摘要 + 子进程 env 注入 + 循环导入修复 
源码：
- runtime/__init__.py: resume_jobs/resume_service 改为懒加载，打破循环导入
- freshrss_pipeline_jobs.py / resume_jobs.py: 子进程注入 .env 环境变量
- freshrss_pipeline.py: LLM 摘要串行改并行 (ThreadPoolExecutor, max_workers=4)

配置：
- term_aliases: 19→149 条，大幅扩充别名映射
- term_stopwords: 19→132 条，增加过滤规则
- filter_context.personal.json: +7 个兴趣关键词
 2026-07-28 16:24:31 +08:00
root 590d050218 keyword cleanup: v2 engine, alias rule layer, LLM semantic suggestions 
- build_review_bundle.py: 新增 _compute_percentile/_compute_growth,
  候选池从固定阈值改为百分位排名 + 增速因子 (v2 policy)
- term_cleanup_policy.json: 升级 v2 schema
- generate_term_cleanup_suggestions.py: 新增 _prepare_alias_suggestions,
  规则层输出 alias (大小写/单复数/分词变体)
- generate_term_cleanup_semantic_suggestions.py: 新增 LLM 语义建议脚本
  (DeepSeek API, 产出 semantic alias/stopword/promote)
- SKILL.md: 更新为 5 Phase 工作流程
- 首轮清洗 apply: interest 54, aliases 17组, stopwords 17个
- docs/design/keyword-cleanup-flow-overview.md: 流程文档
- plans/: 引擎设计方案
 2026-05-14 17:17:49 +08:00
root 4399c9ca90 feat: digest-brief now includes review candidates (not just keep) 2026-04-28 10:39:54 +08:00
zhuyongxin 8136301ad4 fix(runtime): async article summary job startup to prevent MCP timeout 
Root cause:
- RunStore.save() called 4 times synchronously before returning MCP response
- MCP stdio transport blocked by disk I/O and file locks
- Client timed out (-32001) even though job actually started

Fix:
- Use ThreadPoolExecutor to launch job in background thread
- Main thread returns immediately after creating job directory and input file
- Background thread handles subprocess.Popen and finish_stage persistence

Reference: issues/mcp-timeout-job-running.md
 2026-04-16 18:37:13 +08:00
root b29cd8f934 docs: add issue - MCP start_article_summary_job timeout but job actually runs 2026-04-16 11:22:04 +08:00
root 4c4d1a45e6 docs: narrow keyword cleanup skill boundaries 2026-04-14 17:09:23 +08:00
root b8727f1885 feat: add async resume jobs and doc navigation 2026-04-14 15:55:02 +08:00
root 6705613aa4 fix(reader): include all keep candidates in digest brief 2026-04-14 10:10:25 +08:00
root 416414ae1d chore: normalize term stopwords ordering 2026-04-14 09:51:57 +08:00
root f3e7488fc8 docs: add reference templates for IMA notes and public digest 2026-04-14 09:51:57 +08:00
root e3663f681d feat: add keyword cleanup docs, skill updates, and delivery compatibility fix 2026-04-14 09:51:57 +08:00
root a06f2a1d08 chore: update term configs, skill docs and Python 3.10 compat 
- term_change_log: record watch term add history
- term_watchlist: add initial watch terms from cleanup review
- term_aliases: minor update
- filter_context.personal: reorganize interest keywords
- keyword-cleanup-review skill: clarify JSON as formal artifact, markdown as temp review copy
- openclaw_delivery: add Python <3.11 UTC import compat
- daily-keyword-index-design: update suggestion artifact semantics
 2026-04-14 09:51:57 +08:00
zhuyongxin c0194647d8 feat: add sort params 2026-04-13 17:42:39 +08:00
root c528e0abc7 fix(article-summary): 修复 article_summary pipeline 的 3 个 bug 
1. Bug 3 (文件名含斜杠导致路径错误):
   - safe_title 生成时用 .replace('/', '-') 处理 '/' 字符
   - 同时清理连续 dash (--+) 和首尾 dash
   - 原本只处理空格，导致含 '/' 的标题写出路径错误

2. Bug 2 (delivery payload 格式不匹配):
   - 新增 '_iter_selected_items' 对 'candidates' 数组格式的支持
   - 自动 normalize selected_ids 的 'cand:' 前缀
   - candidate 条目同时支持 item_id 字段(新增)和 candidate_id(兼容)
   - 传入 delivery payload + candidate_id 时可正常匹配

3. Pipeline 补充 item_id 字段:
   - OpenClawCandidateInput 加 item_id 字段
   - build_openclaw_candidate_input 填充 item_id
   - 使得后续 article-summary 可通过 item_id 关联 extracted 文件
 2026-04-13 17:38:40 +08:00
root 52ce6bfdf5 fix: validate article summary extracted path input 2026-04-11 23:04:13 +08:00
root c58d8114cd feat: add async job entrypoint for freshrss pipeline 2026-04-11 22:12:11 +08:00
root 10f9cd088f fix: fallback to repo .env for reader pipeline settings 2026-04-11 16:29:19 +08:00
root 2053cccfef feat(reader): add async article summary jobs 2026-04-10 16:37:43 +08:00
root 87d18e4263 feat: harden article summary timeout and fallback 2026-04-07 21:08:06 +08:00
root 2df0af5b82 docs: add openclaw orchestration flow for reader MCP 2026-04-07 15:57:39 +08:00
root 4f219ef92f docs: formalize reader MCP workflow service handoff 2026-04-07 15:32:46 +08:00
root c622bc6247 Implement minimal resume_run for freshrss runs 2026-04-07 15:19:44 +08:00
root d91cdbc6c9 docs: narrow resume_run minimal recovery design 2026-04-07 15:00:19 +08:00
root 4a02894c43 Add MCP delivery payload and run report queries 2026-04-07 13:49:06 +08:00
root d9173fbe0f feat: add MCP run status query tools 2026-04-07 11:17:05 +08:00
root 7563aa8fea docs: add reader MCP architecture and implementation plan 2026-04-07 10:57:46 +08:00
root 72a6853c03 Add run-state persistence for FreshRSS pipeline 2026-04-07 10:48:00 +08:00
root 3a85d47f00 docs: record formal digest artifact workflow 2026-04-01 18:32:25 +08:00
root b5ec58cd6a feat: add public digest brief output for openclaw 2026-04-01 16:48:03 +08:00
root 5b51332075 docs: clarify openclaw handoff boundaries 2026-03-31 18:55:15 +08:00
root f72c9ad2b1 docs: translate README to Chinese 2026-03-31 16:05:50 +08:00
root 660f03674b docs: align reader docs with digest SOP defaults 2026-03-31 15:55:15 +08:00
wdm ca86705646 Translate skill docs (article-deep-summary, llm-summary-review, keyword-cleanup-review) from English to Chinese 2026-03-30 23:14:22 +08:00
zhuyongxin 5b8df317ef Add article-deep-summary skill and enrich article summary prompt 
- Create skills/article-deep-summary/SKILL.md and agents/openai.yaml
- Relax prompt constraints to allow 2-4 sentence elaboration per field
- Add validation output for reference article
- Update plan checklist as completed
 2026-03-30 18:44:30 +08:00
zhuyongxin e24db99baa Add plan for article-deep-summary skill 2026-03-30 17:04:24 +08:00
wdm 1edbc2b736 Remove __pycache__ files from git tracking 2026-03-30 01:12:18 +08:00
wdm 9aa833d2c8 Move CODE_REVIEW.md to plans/ directory 2026-03-30 01:10:05 +08:00
wdm 40fccbd0cc Update README: add generate_article_summaries tool and single-item format support 2026-03-30 01:06:17 +08:00
wdm d3417e29e1 Support single-item extracted file format in summarize_selected_articles 2026-03-30 01:01:43 +08:00
wdm 8575033528 Relax summary field max_length from 140 to 300 characters 2026-03-29 17:58:19 +08:00
wdm ed0e3a9c7d feat: make article summary prompt independent from daily digest 
- Add outputs/prompts/article-summary-prompt.txt with deep knowledge
  note schema (core_conclusion, main_argument, key_methods, etc.)
- Add ArticleSummaryResult Pydantic model for the new schema
- Add validate_article_summary_payload() validator
- Add optional validator param to run_loop_payload(); defaults to
  existing validate_llm_result_payload so daily digest is unaffected
- Update article_summary.py: new prompt path, new validator, new
  Markdown output with Chinese section headers
 2026-03-29 16:49:05 +08:00
wdm 79509fd1ba Add detailed step comments to _process_item helper 2026-03-29 16:27:11 +08:00
wdm 3d58408687 Refactor: split run_freshrss_pipeline into internal helpers 
Extract three internal helpers to reduce the main function from ~330
lines to ~80 lines:
- _process_item(): handles per-item extract/summarize/filter/build
- _build_and_persist_delivery(): builds payload and persists keyword index
- _build_run_report(): assembles status counts and run report dict

No behavior changes. External signature and return shape unchanged.
 2026-03-29 16:27:11 +08:00
wdm 78039eff47 Document drop rule constraint: stop_on_match must be true 
Add explicit authoring constraint in filter-rule-engine-usage.md:
all drop rules must set stop_on_match=true to prevent low-priority
drop rules from overriding high-priority keep decisions.
 2026-03-29 16:27:11 +08:00
wdm a2f6f8a28d Improve _load_required_env error message clarity 
Distinguish between missing argument and missing environment variable
in the RuntimeError message to speed up debugging.
 2026-03-29 16:27:11 +08:00