Search
By AMD-AGI
Skills
Sort:BestMost starsTrending todayTrending this weekTrending this monthNewestRecently updatedName
| # | Skill | Repository | Stars | Used in | Tokens | Auto-check | Licence | Updated |
|---|---|---|---|---|---|---|---|---|
| 1 | Critic layer for the inference optimizer. An agent skill from AMD-AGI/Hyperloom. | AMD-AGI/ | 219 | — | ~3.4k | Automated safety check: Pass | Unknown | today |
| 2 | Inspect the current Spur (AMD SLURM-compatible) cluster node-allocation state and produce a Markdown report (optionally rendered to a standalone HTML page) covering the caller's account/QoS… | AMD-AGI/ | 131 | — | ~2.7k | Automated safety check: Pass | Unknown | today |
| 3 | Review a Hyperloom pull request. An agent skill from AMD-AGI/Hyperloom. | AMD-AGI/ | 219 | — | ~4.8k | Automated safety check: Pass | Unknown | today |
| 4 | Compare a Primus backend against an upstream repository or reference, verify git state, dependencies, directory changes, and integration coupling, then generate comparison reports, dashboard… | AMD-AGI/ | 131 | — | ~3.4k | Automated safety check: Pass | Unknown | today |
| 5 | Deploys or restarts the shared global Experience KB that Hyperloom workspaces push their Experiences to and pull others' from, validates authenticated health, and tells each workspace which .env… | AMD-AGI/ | 219 | — | ~1.1k | Automated safety check: Notes | Unknown | today |
| 6 | Inventory and explain the patch (monkey-patch) optimizations Primus layers over upstream training backends such as Megatron-LM, TorchTitan, and MaxText, including their version compatibility… | AMD-AGI/ | 131 | — | ~3.5k | Automated safety check: Pass | Unknown | today |
| 7 | Run, monitor, stop and report Primus convergence tests -- training a model on a real corpus and checking that the loss curve is healthy -- from a plain-language request such as "run convergence test… | AMD-AGI/ | 131 | — | ~2.1k | Automated safety check: Pass | Unknown | today |
| 8 | Update the Primus documentation for a monthly training-image release — bump version references, realign the bare-metal install scripts with the release Dockerfiles, record the published image stacks… | AMD-AGI/ | 131 | — | ~3.1k | Automated safety check: Pass | Unknown | today |
| 9 | Run a 4-hour multi-node Hyperloom Qwen3-30B-A3B optimization (Infera PD-disaggregated or RayJob aggregated) with --nodes 2 and sglang MoE tuning on MI325X. | AMD-AGI/ | 219 | — | ~1.8k | Automated safety check: Pass | Unknown | today |
| 10 | Validate SLURM cluster nodes by running actual training jobs in groups. | AMD-AGI/ | 131 | — | ~3.7k | Automated safety check: Pass | Unknown | today |
| 11 | Generate a validation plan and test scaffolding to check a Primus optimization that has been ported into a user's own training framework, across correctness (numerical accuracy vs a reference)… | AMD-AGI/ | 131 | — | ~1.5k | Automated safety check: Pass | Unknown | today |
| 12 | Check available idle nodes in a SLURM cluster. An agent skill from AMD-AGI/Primus. | AMD-AGI/ | 131 | — | ~3k | Automated safety check: Warn | Unknown | today |
| 13 | Opinionated guide for using Primus Projection to choose parallelism (TP/PP/EP/CP/DP) and pipeline schedules, validate memory fit on target nodes, reason about communication collectives, and explore… | AMD-AGI/ | 131 | — | ~5.4k | Automated safety check: Pass | Unknown | today |
| 14 | Configures Hyperloom after pip install --target . An agent skill from AMD-AGI/Hyperloom. | AMD-AGI/ | 219 | — | ~7.1k | Automated safety check: Notes | Unknown | today |
