-
Notifications
You must be signed in to change notification settings - Fork 49.3k
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
- Status: Open.#102224 In NousResearch/hermes-agent;
- Status: Open.#102221 In NousResearch/hermes-agent;
[Bug]: /goal stalls after non-failed max_iterations_reached summary in Desktop
comp/agentCore agent runtime: loop, agent_init, prompt builder, context-compression, responses endpointCore agent runtime: loop, agent_init, prompt builder, context-compression, responses endpointcomp/desktopElectron desktop app (apps/desktop/*)Electron desktop app (apps/desktop/*)comp/tuiTerminal UI (ui-tui/ + tui_gateway/)Terminal UI (ui-tui/ + tui_gateway/)P2Medium — degraded but workaround existsMedium — degraded but workaround existssweeper:risk-session-stateSweeper risk: may lose/corrupt/mis-associate session or context stateSweeper risk: may lose/corrupt/mis-associate session or context statetype/bugSomething isn't workingSomething isn't workingStatus: Open.#102213 In NousResearch/hermes-agent;Community report: Hermes 0.21.0 CLI/Desktop running on Intel macOS x86_64 via OCLP
comp/cliCLI entry point, hermes_cli/, setup wizardCLI entry point, hermes_cli/, setup wizardcomp/desktopElectron desktop app (apps/desktop/*)Electron desktop app (apps/desktop/*)P3Low — cosmetic, nice to haveLow — cosmetic, nice to havetype/docsDocumentation improvementsDocumentation improvementsStatus: Open.#102211 In NousResearch/hermes-agent;CLI re-prints the entire streamed response when transform_llm_output makes in-place edits (punctuation normalization)
area/streamingStreaming responses: gateway delivery, provider wireStreaming responses: gateway delivery, provider wirecomp/cliCLI entry point, hermes_cli/, setup wizardCLI entry point, hermes_cli/, setup wizardcomp/pluginsPlugin system and bundled pluginsPlugin system and bundled pluginsP3Low — cosmetic, nice to haveLow — cosmetic, nice to havetype/bugSomething isn't workingSomething isn't workingStatus: Open.#102203 In NousResearch/hermes-agent;[Bug] state.db page 0 clobbered by a write ~0.9s after SIGTERM during graceful shutdown → "file is not a database" on next start
area/sessionsSession lifecycle, resume, persistence, historySession lifecycle, resume, persistence, historycomp/agentCore agent runtime: loop, agent_init, prompt builder, context-compression, responses endpointCore agent runtime: loop, agent_init, prompt builder, context-compression, responses endpointcomp/gatewayGateway runner, session dispatch, deliveryGateway runner, session dispatch, deliveryneeds-reproBug needs reproduction stepsBug needs reproduction stepsP2Medium — degraded but workaround existsMedium — degraded but workaround existssweeper:risk-session-stateSweeper risk: may lose/corrupt/mis-associate session or context stateSweeper risk: may lose/corrupt/mis-associate session or context statetype/bugSomething isn't workingSomething isn't workingStatus: Open.#102198 In NousResearch/hermes-agent;groups.*: no JSON-RPC method exposes per-member live task status inside a hosted room
comp/gatewayGateway runner, session dispatch, deliveryGateway runner, session dispatch, deliverycomp/tuiTerminal UI (ui-tui/ + tui_gateway/)Terminal UI (ui-tui/ + tui_gateway/)needs-decisionAwaiting maintainer decision before any implementationAwaiting maintainer decision before any implementationP3Low — cosmetic, nice to haveLow — cosmetic, nice to havesweeper:risk-session-stateSweeper risk: may lose/corrupt/mis-associate session or context stateSweeper risk: may lose/corrupt/mis-associate session or context statetype/featureNew feature or requestNew feature or requestStatus: Open.#102196 In NousResearch/hermes-agent;[Bug]: CLI path never persists api_content sidecar → injected <memory-context>/tool-result decorations dropped at turn boundary → first API call of every new turn misses prompt cache
comp/agentCore agent runtime: loop, agent_init, prompt builder, context-compression, responses endpointCore agent runtime: loop, agent_init, prompt builder, context-compression, responses endpointP0Critical — data loss, security, crash loopCritical — data loss, security, crash loopsweeper:risk-cachingSweeper risk: may break/degrade prompt caching or cache-key stability (invariant)Sweeper risk: may break/degrade prompt caching or cache-key stability (invariant)sweeper:risk-session-stateSweeper risk: may lose/corrupt/mis-associate session or context stateSweeper risk: may lose/corrupt/mis-associate session or context statetool/memoryMemory tool and memory providersMemory tool and memory providerstype/bugSomething isn't workingSomething isn't workingStatus: Open.#102194 In NousResearch/hermes-agent;[security] hermes update creates root owned files at every update, breaking functionality
area/install-updateInstaller, updater, packaging, wheels, doctorInstaller, updater, packaging, wheels, doctorcomp/cliCLI entry point, hermes_cli/, setup wizardCLI entry point, hermes_cli/, setup wizardneeds-reproBug needs reproduction stepsBug needs reproduction stepsP2Medium — degraded but workaround existsMedium — degraded but workaround existssweeper:risk-compatibilitySweeper risk: may break existing users, config, migrations, defaults, or upgradesSweeper risk: may break existing users, config, migrations, defaults, or upgradestype/bugSomething isn't workingSomething isn't workingStatus: Open.#102193 In NousResearch/hermes-agent;feat(observability): per-conversation execution records with usage attribution
area/usage-costToken accounting, usage reporting, billing, cost trackingToken accounting, usage reporting, billing, cost trackingcomp/agentCore agent runtime: loop, agent_init, prompt builder, context-compression, responses endpointCore agent runtime: loop, agent_init, prompt builder, context-compression, responses endpointP3Low — cosmetic, nice to haveLow — cosmetic, nice to havetype/featureNew feature or requestNew feature or requestStatus: Open.#102190 In NousResearch/hermes-agent;bug(image-routing): recognize valid Codex
-900kvariants when resolving vision capabilitycomp/agentCore agent runtime: loop, agent_init, prompt builder, context-compression, responses endpointCore agent runtime: loop, agent_init, prompt builder, context-compression, responses endpointP2Medium — degraded but workaround existsMedium — degraded but workaround existsprovider/openaiOpenAI / Codex Responses APIOpenAI / Codex Responses APItool/visionVision analysis and image generationVision analysis and image generationtype/bugSomething isn't workingSomething isn't workingStatus: Open.#102189 In NousResearch/hermes-agent;feat(gateway): configurable acknowledgement for captionless image-only messages
area/configConfig system, migrations, profilesConfig system, migrations, profilescomp/gatewayGateway runner, session dispatch, deliveryGateway runner, session dispatch, deliveryP3Low — cosmetic, nice to haveLow — cosmetic, nice to havesweeper:risk-compatibilitySweeper risk: may break existing users, config, migrations, defaults, or upgradesSweeper risk: may break existing users, config, migrations, defaults, or upgradessweeper:risk-message-deliverySweeper risk: may drop, duplicate, misroute, or suppress messagesSweeper risk: may drop, duplicate, misroute, or suppress messagessweeper:risk-session-stateSweeper risk: may lose/corrupt/mis-associate session or context stateSweeper risk: may lose/corrupt/mis-associate session or context statetool/visionVision analysis and image generationVision analysis and image generationtype/featureNew feature or requestNew feature or requestStatus: Open.#102188 In NousResearch/hermes-agent;