{"repo":"NousResearch/hermes-agent","public_id":"7724","display_label":"RL Training docs missing links to related tools","summary":"These issues and PRs address missing hyperlinks to Atropos, Tinker, and Weights & Biases (wandb) in the User Guide's RL Training page. Multiple fix attempts were made to add the absent references.","subsystem":null,"counts":{"members":4,"open_issues":0,"open_issues_distinct":0,"open_prs":0,"closed_issues":1,"merged_prs":1,"attempt_prs":2},"landed_fix":{"status":"merged_to_main","basis":"structural","pr":14506,"merged_at":"2026-04-23T10:07:06+00:00"},"resolution":{"state":"unresolved","by":null},"freshness":{"state":"stale","last_verified_at":"2026-04-23T10:07:06+00:00","is_stale":true},"grouping":{"backbone":"STRUCTURAL_SPANNING","link_density":0.5},"fix_edges":[{"issue_number":7724,"pr_number":14506,"tier":"verified","basis":"superseded_ref","confidence":0.9,"recoverable_from_timeline":false,"pr_merged_at":"2026-04-23T10:07:06+00:00"}],"members":[{"kind":"issue","number":7724,"title":"[Bug]: No links for related tools in User guide - RL Training page","url":null,"state":"closed"},{"kind":"pr","number":7727,"title":"Fix: User guide missing links to Atropos and wandb","url":null,"state":"closed","close_reason":{"refs":[{"kind":"pr","value":"#14506","number":14506}],"category":"superseded_by_pr","implication":"superseded"},"merged":false},{"kind":"pr","number":11749,"title":"docs(rl-training): link Atropos, Tinker, and Weights & Biases","url":null,"state":"closed","close_reason":{"category":"wontfix_bydesign","implication":"wontfix"},"merged":false},{"kind":"pr","number":14506,"title":"docs(rl-training): link Atropos, Tinker, and wandb (salvage #7727)","url":null,"state":"merged","close_reason":{"category":"merged"},"merged":true,"is_fix_pr":true,"fix_tier":"verified","files_changed":["scripts/release.py","website/docs/user-guide/features/rl-training.md"],"fix_hunks":[{"file":"scripts/release.py","hunks":["@@ -402,6 +402,7 @@\n     \"[redacted:email]\": \"TaroballzChen\",\n     \"92638503+Lind3ey@users.noreply.github.com\": \"Lind3ey\",\n     \"[redacted:email]\": \"phpoh\",\n+    \"[redacted:email]\": \"vivganes\",\n }\n \n "]},{"file":"website/docs/user-guide/features/rl-training.md","hunks":["@@ -12,8 +12,8 @@ Hermes Agent includes an integrated RL (Reinforcement Learning) training pipelin\n \n The RL training system consists of three components:\n \n-1. **Atropos** — A trajectory API server that coordinates environment interactions, manages rollout groups, and computes advantages\n-2. **Tinker** — A training service that handles model weights, LoRA training, sampling/inference, and optimizer steps\n+1. **[Atropos](https://github.com/NousResearch/atropos)** — A trajectory API server that coordinates environment interactions, manages rollout groups, and computes advantages\n+2. **[Tinker](https://thinkingmachines.ai/tinker/)** — A training service that handles model weights, LoRA training, sampling/inference, and optimizer steps\n 3. **Environments** — Python classes that define tasks, scoring, and reward functions (e.g., GSM8K math problems)\n \n The agent can discover environments, configure training parameters, launch training runs, and monitor metrics — all through a set of `rl_*` tools.","@@ -24,7 +24,7 @@ RL training requires:\n \n - **Python >= 3.11** (Tinker package requirement)\n - **TINKER_API_KEY** — API key for the Tinker training service\n-- **WANDB_API_KEY** — API key for Weights & Biases metrics tracking\n+- **WANDB_API_KEY** — API key for [Weights & Biases](https://wandb.ai/) metrics tracking\n - The `tinker-atropos` submodule (at `tinker-atropos/` relative to the Hermes root)\n \n ```bash\n"]}]}],"trust":{"default_tier":"verified","includes_suggested":false}}